Intermediate commit.

Ignore unknown engines in --gtid-pos-auto-engines command-line options (but
not SET GLOBAL). This seems useful, to allow a default that auto-creates the
gtid pos table for engines like TokuDB and MyRocks (which greatly benefit
from such), but does not prevent server startup when those engines are not
available.

This was found during MDEV-12179 development, but unrelated so put in its
own commit.

The "restart: --extra-option" facility in expect file for server restart
sets extra options for the restarted mysqld. These options were incorrectly
not cleared when starting a new test case, so the server was restarted with
whatever extra options were left by the previous testcase, causing random
failures depending on order of running tests.

Intermediate commit.

Implement the command-line part of --gtid-pos-auto-engines.

(The option is still not actually used for anything, this will be for
a later commit).

Intermediate commit.

Implement a --gtid-pos-auto-engines system variable. The variable is a list
of engines for which mysql.gtid_slave_pos_ENGINE should be auto-created if
needed.

This commit only implements the option variable. It is not yet used to
actually auto-create any tables, nor is there a corresponding command-line
version of the option yet.

Intermediate commit.

This commit implements that record_gtid() selects a gtid_slave_posXXX table
with a storage engine already in use by current transaction, if any.

The default table mysql.gtid_slave_pos is used if no match can be found on
storage engine, or for GTID position updates with no specific storage
engine.

Table discovery of mysql.gtid_slave_pos* happens on initial GTID state load
as well as on every START SLAVE. Some effort is made to make this possible
without additional locking. New tables are added using lock-free atomics.
Removing tables requires stopping all slaves first. A warning is given in
the error log when a table is removed but a non-stopped slave still has a
reference to it.

If multiple mysql.gtid_slave_posXXX tables with same storage engine exist,
one is chosen arbitrarily to be used, with a warning in the error log. GTID
data from all tables is still read, but only one among redundant tables with
same storage engine will be updated.

Intermediate commit.

When deleting old GTIDs, only pick those GTIDs that were recorded using the
same engine.

(Currently this is not used, there is only one table/engine active. This is
in preparation for extending record_gtid to pick the table with the same
engine as the active transaction, if any).

Intermediate commit.

For each GTID recorded in mysq.gtid_slave_pos, keep track of which
engine the update was made in.

This will be later used to know which rows can be deleted in the table
of a given engine.

Intermediate commit.

Keep track of which mysql.gtid_slave_posXXX tables are available for each
engine, by searching for all tables in the mysql schema with names that
start with "gtid_slave_pos".

The list is computed at server start when the GTID position is loaded, and
it is re-computed on every START SLAVE command. This way, the DBA can
manually add a table for a new engine, and it will be automatically picked
up on next START SLAVE, so a full server restart is not needed.

The list is not yet actually used in the code.

Intermediate commit.

Refactor scan_all_gtid_slave_pos_table() so it can do generic processing on
the list of mysql.gtid_slave_pos_XXX tables.

Intermediate commit.

On server start, look for and read all tables mysql.gtid_slave_pos* to
restore the GTID position.

Simple test case that moves the data to a new
mysql.gtid_slave_pos_innodb table and verifies that the new table is
read at server start.

debug build with older cmake.

The constant is implicitely defined by VS when chosen C runtime is Debug
(/MTd, MDd).

CMake does not define it since https://public.kitware.com/Bug/view.php?id=15777
was fixed.

We remove it from compile flags, to be able to build Debug with /MT runtime
using older cmakes.

Also  do not use BUILD_BYPRODUCTS in ExternalPeoject for older cmake

The patch actually fixes the old defect of the optimizer that
could not extract keys for range access from IN predicates
with row arguments.

This problem was resolved in the mysql-5.7 code. The patch
supersedes what was done there:
 - it can build range access when not all components of
 the first row argument are refer to the columns of the table
 for which the range access is constructed.
 - it can use equality predicates to build range access
 to the table that is not referred to in this argument.

Also, some MDEV-11738/MDEV-11581 post-push fixes.

In MariaDB 10.1, there is no fil_space_t::is_being_truncated field,
and the predicates fil_space_t::stop_new_ops and fil_space_t::is_stopping()
are interchangeable. I requested the fil_space_t::is_stopping() to be added
in the review, but some added checks for fil_space_t::stop_new_ops were
not replaced with calls to fil_space_t::is_stopping().

buf_page_decrypt_after_read(): In this low-level I/O operation, we must
look up the tablespace if it exists, even though future I/O operations
have been blocked on it due to a pending DDL operation, such as DROP TABLE
or TRUNCATE TABLE or other table-rebuilding operations (ALTER, OPTIMIZE).
Pass a parameter to fil_space_acquire_low() telling that we are performing
a low-level I/O operation and the fil_space_t::is_stopping() status should
be ignored.

Met additional issues while trying to enable it.

information_schema. Added .opt file to enable running this
test as a part of innodb_zip suite and recorded current
correct results.

- Also rename plugin's config file
- And fix unit test's CMakeLists.txt

Copy the detection logic from 'rocksdb' test suite, add a note.

Merge branch 'bb-10.2-mariarocks-wlad' of https://github.com/mariadb/server into bb-10.2-mariarocks-wlad

Default to static release (previously static debug was used in debug
builds, but not is appears to be too slow)

Restore MYSQL_SYSVAR_UINT64_T and MYSQL_SYSVAR_SIZE_T in
plugin.h

They use much more disk space than all other tests in rocksdb test suite

Don't use Perl autodie module as there are platforms where it is not present

fixes many test failures in optimized builds in --ps-protocol

vcols in the table definition are intentionaly bad and produce
warnings when a table is opened. In some cases (depending on the
TDC/TC state as left by earlier tests) a table might be opened in the
middle of the test, resulting in spurious warnings. Suppress them.

Also, don't run main.mdev-504 for ps-protocol at all (instead of
disabling ps-protocol inside the test, but still running the test)