[Enhancement] Concurrent Snapshots for Replicated Subscriptions controller could cause BadVersionException

### Search before reporting

- [x] I searched in the [issues](https://github.com/apache/pulsar/issues) and found nothing similar.

### Motivation

Replicated subscription snapshot creation is scheduled at a fixed interval. There is a risk of initiating overlapping snapshot tasks in slow environments with small interval.

This could result in the following exception in the ManagedLedger.

> java.util.concurrent.CompletionException: org.apache.bookkeeper.mledger.ManagedLedgerException$BadVersionException: org.apache.pulsar.metadata.api.MetadataStoreException$BadVersionException: org.apache.zookeeper.KeeperException$BadVersionException: KeeperErrorCode = BadVersion for

### Solution

Added a concurrency guard using an AtomicBoolean to ensure only one startNewSnapshot() runs at a time. When a snapshot is already in progress, subsequent attempts are skipped with a warning log.

### Alternatives

_No response_

### Anything else?

_No response_

### Are you willing to submit a PR?

- [x] I'm willing to submit a PR!

Start at the replicated subscriptions controller and trace the scheduled snapshot path into startNewSnapshot(). Check how overlapping snapshot tasks can reach ManagedLedger and where a concurrency guard belongs. Done means concurrent attempts are skipped with a warning and the BadVersionException scenario is covered by the relevant existing tests.

Written by the indexing model from the issue text.