Improve --safe-slave-backup to support MTS

Current `--safe-slave-backup' behaviour:

1. Stop slave SQL thread
2. If `slave_open_temp_tables' != 0 resume slave SQL thread, wait
   for some time and repeat from step 1
3. Complete the backup
4. Resume slave SQL thread

In case when `slave_parallel_workers > 1' and server doesn't use
`MASTER_AUTO_POSITION' add step 1.5:

1.5. Start slave SQL thread with `START SLAVE UNTIL SQL_AFTER_MTS_GAPS',
       wait for slave SQL thread stopped.

Make sure that `safe-slave-backup-timeout' respected.

[1]
https://www.percona.com/blog/2015/01/29/multi-threaded-replication-with-mysql-5-6-use-gtids/