fix: fix stagedUpdateRun PolicyObservedClusterCount and ApprovalRequest name uniqueness

[jwtty]

Description of your changes

1. When scheduling policy is PickAll, we update PolicyObservedClusterCount to the number of scheduled/bound clusterResourceBindings, instead of -1. For PickFix and PickN, we validate if the number of bindings match the expected count.
2. Add a random appendix to the name of ClusterApprovalRequest generated to avoid conflicts.
3. Add more output in one of the integration test to help debugging UT flakes.

Fixes #

I have:

• Run make reviewable to ensure this PR is ready for review.

How has this code been tested

Special notes for your reviewer