Add an fsck style cluster diagnostic utility

**Motivation:** Currently, Pinot does not have an all-encompassing diagnostic command that an operator can run to assess the health of their cluster. When debugging missing data or cluster desyncs, users have to manually stitch together API calls to figure out what is wrong.

A unified diagnostic utility (similar to an fsck for Pinot) would be highly valuable to help users understand cluster health at a glance, reducing MTTR during incidents.

**Proposed Interface:**
Ideally, this could be integrated as a new command in the existing admin scripts, for example:
`bin/pinot-admin.sh CheckCluster -zkAddress localhost:2181 -clusterName PinotCluster`

_PS: If this is something the community considers would be valuable, I would love to be assigned to this and work on the implementation._

Start by reviewing the existing admin scripts, including bin/pinot-admin.sh, and how current commands access cluster APIs. Define which health and synchronization checks the proposed CheckCluster command must perform, then verify that running it with the shown ZooKeeper address and cluster name produces a useful cluster-health report.

Written by the indexing model from the issue text.