Address feedback from the second code review:
- Make SHOW EXPLAIN code convert the query string before pushing it as a warning text.   

- Address feedback from the second code review.

- Make THD::check_killed() an inline function which makes calls to non-inline 
  functions only whern there are APC requests to be served.

- switch SHOW EXPLAIN to using an INFORMATION_SCHEMA table.

- Make SHOW EXPLAIN command be KILLable with KILL QUERY.

- Better comments
- Make unittest compile on Windows

- Fix the year in Monty Program Ab copyrights in the new files.
- Fix permissions handling so that SHOW EXPLAIN's handling is the 
  same as SHOW PROCESSLIST's.

my_apc-t.cc: make it compile on Windows.

- Move standalone tests to a unittest.
- Added comments.

- Add Monty Program Ab copyright in new files
- Change Apc_target::make_apc_call() to accept a C++-style
  functor (instead of C-style function + parameter)

- Remove out-of-date comments, add dbug assertions.

- SHOW EXPLAIN actually produced correct plan
- Apply fix for lp:1013343 to make EXPLAIN and SHOW EXPLAIN uniform.

- Merge with current 5.5-main

- Add the VIA_SYM token into keyword_sp list, which makes it allowed for
  use as keyword and SP label.

Fixed MDEV-331: last_insert_id() returns a signed number

mysql-test/r/auto_increment.result:
  Added test case
mysql-test/t/auto_increment.test:
  Added test case
sql/item_func.h:
  Changed last_insert_id() to be unsigned.

The --debug-no-sync incorrectly defaulted to ON, disabling sync calls
by default which can loose data or cause corruption. Also, the code
used fsync() instead of the sometimes more efficient fdatasync().

- Make SHOW EXPLAIN code correctly print fake_select_lex: both in the case where 
  it has not yet been executed, and when it has been executed.

MDEV-324: SHOW EXPLAIN: Plan produced by SHOW EXPLAIN for a query with TEMPTABLE view loses 'DERIVED' line
- Make SHOW EXPLAIN code take into account that st_select_lex object without joins can be
  a full-featured SELECTs which were already executed and cleaned up.

   INDEX and SHOW EXPLAIN in parallel
- Rework locking code to use the LOCK_thd_data mutex for all synchronization. This also
  fixed MDEV-301.

Analysis:

When the method JOIN::choose_subquery_plan() decided to apply
the IN-TO-EXISTS strategy, it set the unit and select_lex
uncacheable flag to UNCACHEABLE_DEPENDENT_INJECTED unconditionally.
As result, even if IN-TO-EXISTS injected non-correlated predicates,
the subquery was still treated as correlated.

Solution:
Set the subquery as correlated only if the injected predicate(s) depend
on the outer query.

remove the offending assert.
take the test case from mysql Bug#58015

- Make SHOW EXPLAIN ignore range accesses when printing "Range checked for each record" plans.

MDEV-305: SHOW EXPLAIN: ref returned by SHOW EXPLAIN is different from the normal EXPLAIN ('const' vs empty string)
- The problem was that create_ref_for_key() would act differently, depending on 
  whether we're running EXPLAIN or the actual query.
- As the first step, fixed the EXPLAIN printout not to depend on actions in create_ref_for_key().