MDEV-8875 Wrong metadata for MAX(CAST(time_column AS DATETIME))

Item_func_hybrid_field_type did not return correct field_type(), cmp_type()
and result_type() in some cases, because cached_result_type and
cached_field_type were set in independent pieces of the code and
did not properly match to each other.
Fix:
- Removing Item_func_hybrid_result_type
- Deriving Item_func_hybrid_field_type directly from Item_func
- Introducing a new class Type_handler which guarantees that
  field_type(), cmp_type() and result_type() are always properly synchronized
  and using the new class in Item_func_hybrid_field_type.

Added error handling to crash recovery so that we stop instead of
asserting.

Alloc creating crypt_data to empty page at crash recovery and redo.

MDEV-8852 Implicit or explicit CAST from MAX(string) to INT,DOUBLE,DECIMAL does not produce warnings

Added per-table boolean IETF_QUOTES variable to CSV storage engine. It allows to
enable IETF-compatible parsing of embedded quote and comma characters. Disabled
by default.

This patch is based on Percona revision:
https://github.com/percona/percona-server/commit/b32fbf02766f8db00f9ff4e6dc05a1df97c10ff7

Note that original patch adds server variable, while this patch adds per-table
variable.

Addendum:
* Before calling THD::init_for_queries(), flip the current_thd to wsrep
thread so that memory gets allocated for the right THD.
* Use wsrep_creating_startup_threads instead of plugins_are_initialized
as the condition for the execution of THD::init_for_queries() within
start_wsrep_THD(), as use of latter could still leave some room for
race.

Problem:
When mysqld starts as a galera node, it creates 2 system threads
(applier & rollbacker) using start_wsrep_THD(). These threads are
created before plugin initialization (plugin_init()) for SST methods
like rsync and xtrabackup.

The threads' initialization itself can proceed in parallel to mysqld's
main thread of execution. As a result, the thread initialization code
(start_wsrep_THD()) can end up accessing some un/partially initialized
structures (like maria_hton, in this particular case) resulting in
segfault.

Solution:
Fixed by calling THD::init_for_queries() (which accesses maria_hton)
only after the plugins have been initialized.

MDEV-8819: Failing assertion: block->page.space == page_get_space_id(page_align(ptr)) in file buf0buf.cc line 2551

Add error handling on page reading and do not try to access empty pages.

Merge downstream Debian packaging (MDEV-6284)

The fix for MDEV-8466 earlier fixed MDEV-8635 as well. Adding a test only.

- Add EXPLAIN/ANALYZE FORMAT=JSON handling for a few special cases.

introduced by the patch for MDEV-8466 and MDEV-8468.
Using --enable_prepare_warnings and --disable_prepare_warnings
around the affected query, to have the same warning in regular
and --ps mtr runs.

MDEV-8660 TIME(int_zerofill_column) returns a wrong result

MDEV-8832: Encryption meta data should not be stored when ENCRYPTED=DEFAULT and innodb-encrypt-tables=OFF

Item_string::clone_item() creates a new Item_string that
points exactly to the same buffer that the original one does.
Later, Item_string::print() uses c_ptr() for the original Item_string,
which reallocs the original buffer, and the clone remain with
the old freed buffer.
Refactoring the code not to use c_ptr() in Item_string::print().

Folloup: Made encryption rules too strict (and incorrect). Allow creating
table with ENCRYPTED=OFF with all values of ENCRYPTION_KEY_ID but create
warning that nondefault values are ignored. Allow creating table with
ENCRYPTED=DEFAULT if used key_id is found from key file (there was
bug on this) and give error if key_id is not found.

Failed attempt to load dynamic library with STB_GNU_UNIQUE symbol may corrupt
internal dynamic linker data sructures. This looks like libdl bug:
https://sourceware.org/bugzilla/show_bug.cgi?id=14577

In embedded mode "show plugins soname" fails to dlopen ha_mroonga.so (which
has STB_GNU_UNIQUE) due to undefined reference to wsrep_forced_binlog_format.
It crashes later in subsequent call to dlopen() from  pthread_exit().

Disabled plugins.show_all_plugins until dlopen() bug is fixed.

Analysis: Problem sees to be the fact that we allow creating or altering
table to use encryption_key_id that does not exists in case where
original table is not encrypted currently. Secondly we should not
do key rotation to tables that are not encrypted or tablespaces
that can't be found from tablespace cache.

Fix: Do not allow creating unencrypted table with nondefault encryption key
and do not rotate tablespaces that are not encrypted (FIL_SPACE_ENCRYPTION_OFF)
or can't be found from tablespace cache.

Added decompression and after page update recompression support for
import.

Post fix: initialize a variable

and collations from the global name space into private and protected
methods in Item_func_or_sum.

The crash was caused by range optimizer using RANGE_OPT_PARAM::min_key
(and max_key) to store keys. Buffer size was a good upper bound for
range analysis and partition pruning, but not for EITS selectivity
calculations.

Fixed by making these buffers variable-size. The sizes are calculated
from [pseudo]indexes used for range analysis.

Using a CHARSET_INFO pointer instead.

pointer instread, as the "derivation" and "repertoire" parts of
DTCollation were not really used by user_var_entry.

Moving agg_arg_charsets_for_string_result() and
agg_arg_charsets_for_string_result_with_comparison() inside
"protected" section in Item_func_or_sum.