Fix meta tasks used with --flush-cache (#71311) (#71359)

* Remove incorrect code

(cherry picked from commit 88bfc7977d)

changelogs/fragments/fix_meta_tasks_with_flush_cache.yml

@@ -0,0 +1,2 @@
+bugfixes:
+  - Fix execution of the meta tasks 'clear_facts', 'clear_host_errors', 'end_play', 'end_host', and 'reset_connection' when the CLI flag '--flush-cache' is provided.

lib/ansible/plugins/strategy/__init__.py

@@ -182,7 +182,6 @@ class StrategyBase:
         self._final_q = tqm._final_q
         self._step = context.CLIARGS.get('step', False)
         self._diff = context.CLIARGS.get('diff', False)
-        self.flush_cache = context.CLIARGS.get('flush_cache', False)
 
         # the task cache is a dictionary of tuples of (host.name, task._uuid)
         # used to find the original task object of in-flight tasks and to store
@@ -1103,6 +1102,7 @@ class StrategyBase:
 
         skipped = False
         msg = ''
+        # The top-level conditions should only compare meta_action
         if meta_action == 'noop':
             # FIXME: issue a callback for the noop here?
             if task.when:
@@ -1115,7 +1115,7 @@ class StrategyBase:
             self.run_handlers(iterator, play_context)
             self._flushed_hosts[target_host] = False
             msg = "ran handlers"
-        elif meta_action == 'refresh_inventory' or self.flush_cache:
+        elif meta_action == 'refresh_inventory':
             if task.when:
                 self._cond_not_supported_warn(meta_action)
             self._inventory.refresh_inventory()