MDEV-18648: slave_parallel_mode= optimistic default in 10.5

Description:
============
To change 'CONSERVATIVE' @@global.slave_parallel_mode default to 'OPTIMISTIC'
in 10.5.

@sql/sys_vars.cc
Changed default parallel_mode to 'OPTIMISTIC'

@sql/rpl_filter.cc
Changed default parallel_mode to 'OPTIMISTIC'

@sql/mysqld.cc
Removed the initialization of 'SLAVE_PARALLEL_CONSERVATIVE' to
'opt_slave_parallel_mode' variable.

@mysql-test/suite/rpl/t/rpl_parallel_mdev6589.test
@mysql-test/suite/rpl/t/rpl_mdev6386.test
Added 'mtr' suppression to ignore 'ER_PRIOR_COMMIT_FAILED'. In case of
'OPTIMISTIC' mode if a transaction gets killed during "wait_for_prior_commit"
it results in above error "1964". Hence suppression needs to be added for this
error.

@mysql-test/suite/rpl/t/rpl_parallel_conflicts.test
Test has a 'slave.opt' which explicitly sets slave_parallel_mode to
'conservative'. When the test ends this mode conflicts with new default mode.
Hence check test case reports an error. The 'slave.opt' is removed and options
are set and reset within test.

@mysql-test/suite/multi_source/info_logs.result
@mysql-test/suite/multi_source/reset_slave.result
@mysql-test/suite/multi_source/simple.result
Result content mismatch in "show slave status" output. This is expected as new
slave_parallel_mode='OPTIMISTIC'.

@mysql-test/include/check-testcase.test
Updated default 'slave_parallel_mode' to 'optimistic'.

Refactored rpl_parallel.test into following test cases.
Test case 1: @mysql-test/suite/rpl/t/rpl_parallel_domain.test
Test case 2: @mysql-test/suite/rpl/t/rpl_parallel_domain_slave_single_grp.test
Test case 3: @mysql-test/suite/rpl/t/rpl_parallel_single_grpcmt.test
Test case 4: @mysql-test/suite/rpl/t/rpl_parallel_stop_slave.test
Test case 5: @mysql-test/suite/rpl/t/rpl_parallel_slave_bgc_kill.test
Test case 6: @mysql-test/suite/rpl/t/rpl_parallel_gco_wait_kill.test
Test case 7: @mysql-test/suite/rpl/t/rpl_parallel_free_deferred_event.test
Test case 8: @mysql-test/suite/rpl/t/rpl_parallel_missed_error_handling.test
Test case 9: @mysql-test/suite/rpl/t/rpl_parallel_innodb_lock_conflict.test
Test case 10: @mysql-test/suite/rpl/t/rpl_parallel_gtid_slave_pos_update_fail.test
Test case 11: @mysql-test/suite/rpl/t/rpl_parallel_wrong_exec_master_pos.test
Test case 12: @mysql-test/suite/rpl/t/rpl_parallel_partial_binlog_trans.test
Test case 13: @mysql-test/suite/rpl/t/rpl_parallel_ignore_error_on_rotate.test
Test case 14: @mysql-test/suite/rpl/t/rpl_parallel_wrong_binlog_order.test
Test case 15: @mysql-test/suite/rpl/t/rpl_parallel_incorrect_relay_pos.test
Test case 16: @mysql-test/suite/rpl/t/rpl_parallel_retry_deadlock.test
Test case 17: @mysql-test/suite/rpl/t/rpl_parallel_deadlock_corrupt_binlog.test
Test case 18: @mysql-test/suite/rpl/t/rpl_parallel_mode.test
Test case 19: @mysql-test/suite/rpl/t/rpl_parallel_analyze_table_hang.test
Test case 20: @mysql-test/suite/rpl/t/rpl_parallel_record_gtid_wakeup.test
Test case 21: @mysql-test/suite/rpl/t/rpl_parallel_stop_on_con_kill.test
Test case 22: @mysql-test/suite/rpl/t/rpl_parallel_rollback_assert.test

mysql-test/suite/rpl/include/rpl_parallel_analyze_table_hang.inc

@@ -0,0 +1,73 @@
+--echo *** MDEV-7888: ANALYZE TABLE does wakeup_subsequent_commits(), causing wrong binlog order and parallel replication hang ***
+--source include/have_innodb.inc
+--source include/have_debug.inc
+--source include/have_debug_sync.inc
+--source include/master-slave.inc
+
+--connection server_2
+SET @old_parallel_threads=@@GLOBAL.slave_parallel_threads;
+SET @old_parallel_mode=@@GLOBAL.slave_parallel_mode;
+--source include/stop_slave.inc
+SET GLOBAL slave_parallel_mode='conservative';
+SET GLOBAL slave_parallel_threads=10;
+CHANGE MASTER TO master_use_gtid=slave_pos;
+--source include/start_slave.inc
+
+--connection server_1
+ALTER TABLE mysql.gtid_slave_pos ENGINE=InnoDB;
+CREATE TABLE t2 (a int PRIMARY KEY) ENGINE=InnoDB;
+CREATE TABLE t3 (a INT PRIMARY KEY, b INT) ENGINE=InnoDB;
+--save_master_pos
+
+--connection server_2
+--sync_with_master
+--source include/stop_slave.inc
+SET @old_dbug= @@GLOBAL.debug_dbug;
+SET GLOBAL debug_dbug= '+d,inject_analyze_table_sleep';
+
+--connection server_1
+# Inject two group commits. The bug was that ANALYZE TABLE would call
+# wakeup_subsequent_commits() too early, allowing the following transaction
+# in the same group to run ahead and binlog and free the GCO. Then we get
+# wrong binlog order and later access freed GCO, which causes lost wakeup
+# of following GCO and thus replication hang.
+# We injected a small sleep in ANALYZE to make the race easier to hit (this
+# can only cause false negatives in versions with the bug, not false positives,
+# so sleep is ok here. And it's in general not possible to trigger reliably
+# the race with debug_sync, since the bugfix makes the race impossible).
+
+SET @old_dbug= @@SESSION.debug_dbug;
+SET SESSION debug_dbug="+d,binlog_force_commit_id";
+
+# Group commit with cid=10000, two event groups.
+SET @commit_id= 10000;
+ANALYZE TABLE t2;
+INSERT INTO t3 VALUES (120, 0);
+
+# Group commit with cid=10001, one event group.
+SET @commit_id= 10001;
+INSERT INTO t3 VALUES (121, 0);
+
+SET SESSION debug_dbug=@old_dbug;
+
+SELECT * FROM t3 WHERE a >= 120 ORDER BY a;
+--source include/save_master_gtid.inc
+
+--connection server_2
+--source include/start_slave.inc
+--source include/sync_with_master_gtid.inc
+
+SELECT * FROM t3 WHERE a >= 120 ORDER BY a;
+
+# Clean up.
+--source include/stop_slave.inc
+SET GLOBAL debug_dbug= @old_dbug;
+SET GLOBAL slave_parallel_threads=@old_parallel_threads;
+SET GLOBAL slave_parallel_mode=@old_parallel_mode;
+--source include/start_slave.inc
+
+--connection server_1
+DROP TABLE t2,t3;
+
+--source include/rpl_end.inc
+