mgr/cephadm: clear CEPHADM_DAEMON_PREPARE_CREATE_FAIL warnings

This is a new warning with the parallel deployment work
(it would have fallen under the general CEPHADM_APPLY_SPEC_FAIL
before) so we need to make sure to clear it along with the
other deployment related health warnings. Otherwise, once it
happens once the warning sticks around forever even if
the issue has been resolved.

Signed-off-by: Adam King <[email protected]>
(cherry picked from commit da14411)

Resolves: rhbz#2372821

Author: adk3798

src/pybind/mgr/cephadm/serve.py

@@ -636,7 +636,7 @@ def _apply_all_services(self) -> bool:
             # and not dependant on other daemon's placements in any way
             specs = [s for s in _specs if not s.placement.count] + [s for s in _specs if s.placement.count]
 
-        for name in ['CEPHADM_APPLY_SPEC_FAIL', 'CEPHADM_DAEMON_PLACE_FAIL']:
+        for name in ['CEPHADM_APPLY_SPEC_FAIL', 'CEPHADM_DAEMON_PLACE_FAIL', 'CEPHADM_DAEMON_PREPARE_CREATE_FAIL']:
             self.mgr.remove_health_warning(name)
         self.mgr.apply_spec_fails = []
         hosts_altered: Set[str] = set()