MDEV-20844 RBR from binary(16) to inet6 fails with error 171: The event was corrupt, leading to illegal data being read

This patch changes the way how INET6 is packed to the RBR binary log:
- from fixed length 16 bytes
- to BINARY(16) compatible variable length style
  with trailing 0x00 byte compression.

This is to make INET6 fully compatible with BINARY(16) in RBR binary logs,
so RBR replication works in this scenarios:

- Old master BINARY(16) -> New slave INET6
- New master INET6      -> Old slave BINARY(16)

A new class StringPack was added to share the code between
Field_string and Field_inet6.

The old code to print partition values was too complicated:
- it created new Items for character set conversion purposes.
- it mixed string conversion and partition error reporting
  in the same code blocks.

Simplifying the code as follows:

- Adding helper methods String::can_be_safely_convert_to() and
  String::append_introducer_and_hex().

- Adding DBUG_EXECUTE_IF("generate_partition_syntax_for_frm",  push_warning...)
  into generate_partition_syntax_for_frm(), to test the PARTITON
  clause written to FRM. Adding test partition_utf8-debug.test for this.

- Removing functions get_cs_converted_part_value_from_string() and
  get_cs_converted_string_value. Changing get_partition_column_description()
  to use Type_handler::partition_field_append_value() instead.
  This makes SHOW CREATE TABLE and SELECT FROM I_S.PARTITIONS
  use the same code path.

- Changing Type_handler::partition_field_append_value() not to
  call convert_charset_partition_constant(), to avoid creating a new Item
  for string conversion pursposes.
  Rewritting the code to use only String methods.

- Removing error reporting code (ER_PARTITION_FUNCTION_IS_NOT_ALLOWED)
  from Type_handler::partition_field_append_value().
  The error is correctly detected and reported on the caller level.
  So error reporting was redundant here.

Also:

- Moving methods Type_handler::partition_field_*() from sql_partition.cc
  to sql_type.cc. This fixes compilation problem with -DPLUGIN_PARTITION=NO,
  earlier introduced by the patch for MDEV-20831.

Removing MariaDB_FUNCTION_COLLECTION_PLUGIN

- Defining MariaDB_FUNCTION_PLUGIN
- Changing the code in /plugins/type_inet/ and /plugins/type_test/
  to use MariaDB_FUNCTION_PLUGIN instead of MariaDB_FUNCTION_COLLECTION_PLUGIN.
- Changing maturity for the INET6 data type plugin from experimental to alpha.

This clause in CREATE TABLE:

  PARTITION BY LIST COLUMNS (inet6column)
    (PARTITION p1 VALUES IN ('::'))

was erroneously written to frm file as:

  PARTITION BY LIST COLUMNS(inet6column)
    (PARTITION p1 VALUES IN (_binary 0x3A3A))

I.e. the text value '::' was converted to HEX representation
and prefixed with _binary.

A simple fix could write `_latin1 0x3A3A` instead of `_binary 0x3A3A`,
but in case of INET6 we don't need neither character set introducers,
nor HEX encoding, because text representation of INET6 values consist
of pure ASCII characters.

So this patch changes the above clause to be printed as:

  PARTITION BY LIST COLUMNS(inet6column)
    (PARTITION p1 VALUES IN ('::'))

Details:

The old code in check_part_field() was not friendly to pluggable data types.
Replacing this function to two new virtual methods in Type_handler:

  virtual bool partition_field_check(const LEX_CSTRING &field_name,
                                     Item *item_expr) const;

  virtual bool partition_field_append_value(String *str,
                                            Item *item_expr,
                                            CHARSET_INFO *field_cs,
                                            partition_value_print_mode_t mode)
                                            const;

so data type plugins can decide whether they need to use character set
introducer and/or hex encoding when printing partition values.

The code erroneously assumed that only Field_str descendants can
store character set information. After adding Field_inet6, it's
not true anymore.

Also, after adding Field_inet6, storing field->charset() become not correct either:
- Field_inet6::charset() return &my_charset_latin1,
  because clients see INET6 as VARCHAR(39).
- Field_inet6::binlog_type_info().m_cs returns &my_charset_bin
  because storage engines see INET6 as BINARY(16).

We need to store &my_charset_bin to the binlog metadata,
so the slave sees INET6 as BINARY(16), like storage engines do,
to make the slave treat the replicated data as binary IPv6 address
representation (rather than text representation).

The correct character set that needs to be stored to the metadata
is already populated to binlog_type_info_array[i].m_cs. So the
fixed code version uses this value rather than field->charset().

Type_handler_inet6::is_val_native_ready() was not overriden and
returned "false" by default, so Item_sum_min_max::update_field()
erroneously went through the min_max_update_str_field() rather than
min_max_update_native_field() execution path. As a result '8888::'
was compared to 'fff::' in string format (rather than INET6 binary format)
and gave "less" instead of "greater".

Adding the forgotten overriding method returning "true".

Ison_objectagg implemented.

post-review fixes:
* test for dependent subqueries
* test for triggers and routines
* disallow INSERT...RETURNING in triggers and stored functions
* don't return anything if INSERT IGNORE ignored an error

when there are nested subqueries, and a field in a subquery is
resolved as an outer reference to a table few levels up, all subqueries
the subquery with a reference and all subqueries up to subquery with
the table must be marked as dependent.

in the text protocol and PS-prepare step it happens in
Item_field::fix_outer_field in a loop that walks contexts
using context->outer_context.

in PS-execute step Item_field->cached_table is set and subqueries
are walked in a PS-only mark_select_range_as_dependent(),
which inconsistently walks SELECT_LEX'es using select_lex->outer_select().

Fix mark_select_range_as_dependent() to walk contexts, not SELECT_LEX'es,
to have the same logic both in prepare and execute steps.

This fixes a crash in main.insert_returning in --ps-protocol

Closes #1384

comments, whitespaces

because internally setup_wild() adjusts select_lex->with_wild directly
anyway, so there is no reason to pretend that the number of '*' may be
anything else but select_lex->with_wild

And don't update select_lex->item_list, because fields can come
from anywhere and don't necessarily have to be copied into select_lex.

remove inconsistent limitation

Show the config file in I_S.SYSTEM_VARIABLES
But only if the user has FILE privilege

change get_one_option() prototype to pass the filename and
not to pass the redundant optid.

make load_defaults() store the file name in the generated option list
using a special marker ---file-marker--- option.

Pick up this filename in handle_options().

Remove ---args-separator---, use ---file-marker--- with an empty file
name instead - this simplifies checks on the caller, only one special
option to recognize.

only my_getopt should use it, because it changes my_getopt's behavior.
If one simply wants to skip the separator - don't ask it to be added
in the first place

process all --defaults* options uniformly,
get rid of special case for --no-defaults and --print-defaults
use realpath instead of blindly concatenating pwd and relative path.

it turns out that practically every single user of handle_options()
used the get_one_option callback. Simplify the code,
make it mandatory, adjust unit tests.

almost all my_getopt settings and callbacks are global variables,
directly assignable to configure my_getopt. Only getopt_get_addr
was using a setter function. Get rid of it, make it a global
directly assignable variable like all other settings.

Also make getopt_compare_strings() static.

This is a remnant of "MySQL Instance Manager", which was removed in
MySQL-5.5.0 and never existed in MariaDB

Remove callback, simplify and optimize the code accordingly.

MDEV-20818 ER_CRASHED_ON_USAGE or Assertion `length <= column->length' failed in write_block_record on temporary table

The patch for `MDEV-20795 CAST(inet6 AS BINARY) returns wrong result`
unintentionally changed what Item_char_typecast::type_handler()
returns. This broke UNIONs with the BINARY() function, as the Aria
engine started to get columns of unexpected data types.

Restoring previous behaviour, to return
  Type_handler::string_type_handler(max_length).

The prototype for Item_handed_func::return_type_handler() has changed
from:
  const Type_handler *return_type_handler() const
to:
  const Type_handler *return_type_handler(const Item_handled_func *) const

Putting "#ifndef NO_EMBEDDED_ACCESS_CHECKS" around these constant declarations:

const uint max_hostname_length= 60;
const uint max_dbname_length= 64;

Suppress warnings when Field_inet6::store() is called from
read_statistics_for_table() and other optimizer related routines.

This patch does for Field_inet6 the same thing with what Monty's
patch previously did for other Field_xxx:

> commit 1bbe8c5e
> Author: Michael Widenius <monty@mariadb.org>
> Date:   Sun Sep 22 04:08:48 2019 +0300
>
>    Proper fix for disabling warnings in read_statistics_for_table().
>    MDEV-20589: Server still crashes in Field::set_warning_truncated_wrong_value

Alas, some meaningful warnings disappeared.

These tests use only HEAP and MyISAM tables.

Adding the "override" keyword to Item_char_typecast_func_handler_inet6_to_binary::type_handler_for_create_select()

Clang fails to compile without it.

Engine get LIMIT/OFFSET info an can it use/reset.

Kill check moved from send_data() methids in its wrapper:
send_data_with_check().

Now both offset and limit are stored and do not chenged during execution
(offset is decreased during processing in versions before 10.5).

(Big part of this changes made by Monty)

rec_init_offsets(): Relax the assertion that was added in
commit 01f45bec
to catch ROW_FORMAT=REDUNDANT records that have fewer fields
than expected.

This assertion would fail when accessing the records of the
built-in InnoDB table SYS_INDEXES. The column MERGE_THRESHOLD
had been effectively instantly added in MariaDB Server 10.2
(and MySQL 5.7), but is_instant() does not hold for that index.
Relax the assertion, so that it will not fail in this case.

MDEV-20813 Assertion failure in buf_flush_init_for_writing() for innodb_immediate_scrub_data_uncompressed=ON

The assertion that was added in
commit c0c003be
to augment the fix of MDEV-20805 turns out to be invalid when
innodb_immediate_scrub_data_uncompressed is enabled.
In this mode, fsp_init_file_page() will be invoked on data pages
that have been freed, causing writes of almost-all-zero pages.

btr_page_free(): Adjust the comment.

buf_flush_init_for_writing(): Disable the assertion with a note
that it should be re-enabled in MDEV-15528.