PXC-5203: Cluster disruption or deadlock with LOCK TABLES/INSTANCE FO…

[jaideepkarande]

…R BACKUP misuse

https://perconadev.atlassian.net/browse/PXC-5203

A session that ran its own DDL/DML while holding LOCK TABLES FOR BACKUP was replicated via TOI before the local backup-lock check rejected it, so peers applied the DDL while the originator failed it locally -- causing a Galera "inconsistency detected" vote and node eviction.

Resolution:
Similar issue was already fixed for "LOCK INSTANCE FOR BACKUP" extended the condition for "LOCK TABLES FOR BACKUP"

[jaideepkarande]

https://pxc.cd.percona.com/view/8.0%20parallel%20MTR/job/pxc-8.x-pipeline-parallel-mtr/848/console

No failure in build #848 is related to the PXC-5203 LOCK TABLES/INSTANCE FOR BACKUP fix.

• Confirmed the Jenkins job built the exact PXC-5203-8.4 commit (b09b523).
• The new test galera.pxc_backup_lock_toi_reject passed in both enc_off and enc_on.
• Of 21 flagged [ fail ] occurrences, 13 were flaky and passed on MTR's automatic retry; 6 persisted after retry, but all are pre-existing, unrelated issues: Galera
  SST/state-transfer/cluster-formation (galera_pc_bootstrap, galera_wan_restart_sst), GTID/binlog timing (rpl_gtids_table_disable_binlog_on_slave,
  galera_as_slave_gtid_replicate_do_db), and a test-infra missing-include-file error (galera_rpl_pxc_strict_mode_mismatch x2).
• All existing backup-lock/FTWRL tests (pxc_ftwrl_backup_lock_deadlock, pxc_ftwrl_try_pause_timeout_queue, galera_lock_table, the rpl_backup tests) passed — no regression.