Optimize flush tables with read lock (FTWRL) to not wait for select's

Part of MDEV-5336 Implement LOCK FOR BACKUP

The idea is that instead of waiting in close_cached_tables() for all
tables to be closed, we instead call flush_tables() that does:
- Flush not used objects in table cache to free memory
- Collect all tables that are open
- Call HA_EXTRA_FLUSH on the objects, to get them into "closed state"
- Added HA_EXTRA_FLUSH support to archive and CSV
- Added multi-user protection to HA_EXTRA_FLUSH in MyISAM and Aria

The benefit compared to old code is:
- FTWRL doesn't have to wait for long running read operations or
  open HANDLER's

mysql-test/main/mdl_sync.test

@@ -4079,9 +4079,10 @@ disconnect con3;
 --echo # MDEV-12620 - set lock_wait_timeout = 1;flush tables with read lock;
 --echo #              lock not released after timeout
 --echo #
+
 CREATE TABLE t1(a INT) ENGINE=InnoDB;
 SET debug_sync='open_tables_after_open_and_process_table SIGNAL ready WAIT_FOR go';
-send SELECT * FROM t1;
+send INSERT INTO t1 values (1);
 
 connect (con1,localhost,root,,);
 SET debug_sync='now WAIT_FOR ready';
@@ -4093,12 +4094,31 @@ SET debug_sync='now SIGNAL go';
 
 connection default;
 reap;
+
+--echo # After MDEV-5536, SELECT will not block FLUSH TABLES
+
+SET debug_sync='RESET';
+SET debug_sync='open_tables_after_open_and_process_table SIGNAL ready WAIT_FOR go';
+send SELECT * FROM t1;
+
+connection con1;
+SET debug_sync='now WAIT_FOR ready';
+# lock_wait_timeout should be 0 in 10.3, so that we don't have to wait at all
+SET lock_wait_timeout=1;
+FLUSH TABLES WITH READ LOCK;
+SET debug_sync='now SIGNAL go';
+
+connection default;
+reap;
+connection con1;
+unlock tables;
+connection default;
+
 SET debug_sync='RESET';
 DROP TABLE t1;
 
 disconnect con1;
 
-
 # Check that all connections opened by test cases in this file are really
 # gone so execution of other tests won't be affected by their presence.
 --source include/wait_until_count_sessions.inc