Description of problem:

When multiple PATCH api requests are sent, the MulticlusterRoleAssignment panics and mistakenly goes into the delete stage, even though PATCH requests never involve delete. There are also multiple conflict retries, even on basic edit api requests.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:

1. Send multiple patch api requests rapidly back to back on the same MulticlusterRoleAssignment
2. Check logs and view panics
3. Check logs and view MRA deletion flow logs

Actual results:

MRA controller panics, goes into the deletion flow by mistake, and has too many conflict retries on basic 1-request api call actions.

Expected results:

MRA controller should not panic. MRA controller should not go into the deletion flow by mistake, unless MRA is actually deleted. MRA controller should not have conflict retries on basic actions.

Additional info: