MDEV-8725: Assertion `!(thd->rgi_slave && thd-> rgi_slave->did_mark_start_commit)' failed in ha_rollback_trans

The assertion is there to catch cases where we rollback while
mark_start_commit() is active. This can allow following event groups
to be replicated too early, causing conflicts.

But in this case, we have an _explicit_ ROLLBACK event in the binlog,
which should not assert.

We fix this by delaying the mark_start_commit() in the explicit
ROLLBACK case. It seems safest to delay this in ROLLBACK case anyway,
and there should be no reason to try to optimise this corner case.

This bug is essentially another variant of MDEV-7458.

If a transaction conflict caused a deadlock kill of T2 in record_gtid()
during commit, the code would do a rollback _before_ running
rgi->unmark_start_commit(). This creates a race where following transactions
could start too early (before T2 has completed its transaction retry). This
in turn could lead to replication failure, if there was a conflict that
caused eg. duplicate key error or similar.

The fix is to remove these rollbacks (in Query_log_event::do_apply_event()
and Xid_log_event::do_apply_event(). They seem out-of-place; code in
log_event.cc generally does not roll back on error, this is handled higher
up.

In addition, because of the extreme difficulty of reproducing bugs like
MDEV-7458 and MDEV-8302, this patch adds some extra precations to try to
detect (in debug builds) or prevent (in release builds) similar bugs.
ha_rollback_trans() will now call unmark_start_commit() if needed (and
assert in debug build when a caller does rollback without unmark first).

We also add an extra check for thd->killed() so that we avoid doing
mark_start_commit() if we already have a pending deadlock kill.

And we add a missing unmark_start_commit() call in the error case, found by
the above assertion.

fix the test case for ps-protocol

MDEV-5977 MariaDB 10.0 is not installable on Trusty when "trusty-updates universe" is in sources.list

fix upgrade[2] tests on trusty and utopic: add missing
conflicts/replaces, client-10.0 should conflict/replace client-5.5.

reformat other conflicts/replaces lines to make them easier
to read and merge

in ha_delete_table()
* only convert ENOENT and HA_ERR_NO_SUCH_TABLE to warnings
* only return real error codes (that is, not ENOENT and
  not HA_ERR_NO_SUCH_TABLE)
* intercept HA_ERR_ROW_IS_REFERENCED to generate backward
  compatible ER_ROW_IS_REFERENCED

in mysql_rm_table_no_locks()
* no special code to handle HA_ERR_ROW_IS_REFERENCED
* no special code to handle ENOENT and HA_ERR_NO_SUCH_TABLE
* return multi-table error ER_BAD_TABLE_ERROR <table list> only
  when there were many errors, not when there were many
  tables to drop (but only one table generated an error)

in cmake tests let's treat clang like gcc (same options,
same builtins) in many cases.

* don't check the compiler when
  * testing for -fvisibility=hidden support
  * testing for HAVE_ABI_CXA_DEMANGLE
  * testing for HAVE_GCC_ATOMIC_BUILTINS
  * when removing options with string(replace)
  * when running ${CC} --version (ignore the error instead)
* run ABI checks for clang
* use "canonical" gcc flags for clang
* fix groonga too

Also:

* add cmake detection for gcc __atomic_* builtins. they might be
  supported (__ATOMIC_SEQ_CST is defined), but not for all operand
  sizes. In particular, 64-bit atomic load is problematic on i386
* cache check results for Windows
* remove the test for HAVE_CXXABI_H (HAVE_ABI_CXA_DEMANGLE is
  suffifient)

* new mysqldump option
* add more engines to the "external data engines" list
* redo the check to be able to print the list of engines in --help

in innobase: compilation error on windows
other changes: perfschema merge followup

only columns really stored to database.

When RENAME TABLE is executed, it apparently does not check whether the engine
is available (unlike ALTER TABLE .. RENAME, which does). It means that if the
engine in question was not loaded on some reason, the table might become
unusable, since the engine won't know about the change.

With this patch RENAME TABLE fails if storage engine is not available.

Put client plugins into -common rpm, not -shared.
Because they're needed for
* all clients that link with shared libmysqlclient (-shared)
* our clients from -client rpm, they're statically linked (-client)
* the server that acts as a replication slave (-server)

MDEV-7871 Tests fail massively on "Assertion `status_var.memory_used == 0'" when run with --ps --embedded.
   As the MF_THREAD_SPECIFIC was introduced to the alloc_root's and
   the prealloc added to the statement::mem_root and statement::result.alloc, we have to adjust
   the embedded server to it. The preallocation was removed for the embedded server as it
   makes no sence for it. The msyqltest should free the statement inside the proper thead to
   make the memory statistics happy.

MDEV-8131 MariaDB does not build on hurd-i386: plugin/auth_dialog/dialog.c:172:20: error: 'RTLD_DEFAULT' undeclared

define _GNU_SORUCE before including dlfcn.h

when --bind-address is not specificed explicitly (or set to '*')
MariaDB tries all wildcard addresses. Print a warning (not an error)
if a socket cannot be created for some of them.

Still print an error if a socket cannot be created for an address
that a user has specified expicitly with --bind-address.

disable IPv6 globally in mysql-test, only use it
in dedicated IPv6 tests (where it is enabled per-test)

in the previous commit for MDEV-8205.

MDEV-8316: debugger aborting because missing DBUG_RETURN or DBUG_VOID_RETURN macro in function "any_slave_sql_running"

Fix a handful of "return" that should be DBUG_RETURN in sql/rpl_mi.cc.

Follow-up patch to temporarily avoid a sporadic failure in the test
rpl.rpl_000011 due to MDEV-8301.

There is a window during thread exit where the global status is
counted incorrectly - the contribution for the exiting thread is
counted twice. The patch for MDEV-8294 made this window visible to the
test case rpl.rpl_000011, causing it to sporadically fail. Temporarily
silence this with a wait for the expected value; can be removed once
MDEV-8294 is fixed.

MDEV-8212 alter table - failing to ADD PRIMARY KEY IF NOT EXISTS when existing index of same as column name.
        The default name for the primary key is rather 'PRIMARY' instead of the indexed column name.

Add progress info on InnoDB/XtraDB row0merge phase. Note that we
do not know exact number of rounds merge sort needs at start thus
also progress report might not be accurate.

#include <m_string.h> where strnlen() is used

Added test case to show that correct error message is printed when
log file size is too small for big blob.

some duplicate code.