Backport pull request #125 from grooverdan/MDEV-8923_innodb_buffer_pool_dump_pct to 10.0

WL#6504 InnoDB buffer pool dump/load enchantments

This patch consists of two parts:

    1. Dump only the hottest N% of the buffer pool(s)
    2. Prevent hogging the server duing BP load

From MySQL - commit b409342c43ce2edb68807100a77001367c7e6b8e

Add testcases for innodb_buffer_pool_dump_pct_basic.

Part of the code authored by Daniel Black

MDEV-9276: MySQL Bug #78754: FK definitions missing from SHOW CREATE TABLE in "innodb_read_only" mode

Fix logic around retrying failed Windows async IO as synchronous IO . os_file_write/read macros were wrong (had wrong number of args), among other things

  modified:   storage/connect/tabjson.cpp

MDEV-7655 SHOW CREATE TABLE returns invalid DDL when using virtual columns along with a table collation

that was mistakenly merged from mysql-5.5.47

(introduces valgrind failures in main.sp, because Field_varstring
columns are created as FIELD_NORMAL and that causes aria to
read bytes between the actual value length and field max length)

that was mistakenly merged from mysql-5.5.47

Make dependency on mysql-common package in debian/control
versioned to force that it is installed together with the other
mariadb-* packages. This ensures that the init.d and systemd
files for MariaDB mysqld are the ones from MariaDB.org repos.

This is useful for developers who run autobake-deb.sh directly
from the source tree as the resulting source tar package will be
90% smaller and all operations regarding it consequently 10x faster.

This should make the provides/breaks/replaces/conflicts lines
work properly when upgrading from mysql-5.6 to mariadb-packages.

Conflicts:

	debian/control

The only difference between the files was the Debian standards
line by one micro version. There was no logical motivation why
there was separate control file for Debian and Ubuntu.

The new debian/control file lives directly at the correct location
and has the correct contents.

Modify autobake.sh to take into account new control file location.

https://github.com/MariaDB/server/pull/56

while according to Storage Engine API column names should be compared
case insensitively. This can cause FRM and InnoDB data dictionary to
go out of sync.

OPTIMIZE TABLE against MyISAM/Aria table may fail and leave stale temporary file
if mysql_file_create() returns 0 file descriptor. This was due to wrong
condition, which considered 0 as failure.

5.5 doesn't have fix for MDEV-5679, thus 0 file descriptor is always occupied
by stdin and this bug is not reproducible.

10.1 has fix for MDEV-8475, which hides this bug.

No test case since it mostly depends on how OS returns file descriptors.

fixed conversion

Called thd_progress_init() from several threads used for FT-index
creation. For FT-indexes, need better way to report progress,
remove current one for them.

removed outdated code which was causing error (no need reassign result for subqueries & Co)

In fact it was error in decimal library (incorrect processing of buffer overflow) invisible from other server parts because of buffer allocation and precision tests.

Correct InnoDB calls to progress report API:
- second argument of thd_progress_init() is number of stages, one stage is
  enough for the row merge
- third argument of thd_progress_report() is some value indicating threshold,
  for the row merge it is file->offset
- second argument of thd_progress_report() is some value indicating current
  state, for the row merge it is file->offset - num_runs.