ClusterManagerTaskThrottler Improvements (opensearch-project#15508)

* [ClusterManagerTaskThrottler Improvements] : Add shallow check in ClusterManagerTaskThrottler to fail fast before computeIfPresent to avoid lock when queue is full

Signed-off-by: Sumit Bansal <sumitsb@amazon.com>

Author: sumitasr

CHANGELOG.md

@@ -44,6 +44,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Add canRemain method to TargetPoolAllocationDecider to move shards from local to remote pool for hot to warm tiering ([#15010](https://github.com/opensearch-project/OpenSearch/pull/15010))
 - Add support for pluggable deciders for concurrent search ([#15363](https://github.com/opensearch-project/OpenSearch/pull/15363))
 - [Remote Publication] Added checksum validation for cluster state behind a cluster setting ([#15218](https://github.com/opensearch-project/OpenSearch/pull/15218))
+- ClusterManagerTaskThrottler Improvements ([#15508](https://github.com/opensearch-project/OpenSearch/pull/15508))
 
 ### Dependencies
 - Bump `netty` from 4.1.111.Final to 4.1.112.Final ([#15081](https://github.com/opensearch-project/OpenSearch/pull/15081))