Do not use a fixed size queue anymore (avoid the thread to be blocked).

Author: afoucret

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/inference/bulk/BulkInferenceExecutor.java

@@ -9,14 +9,14 @@
 
 import org.elasticsearch.action.ActionListener;
 import org.elasticsearch.common.util.concurrent.AbstractRunnable;
+import org.elasticsearch.common.util.concurrent.ConcurrentCollections;
 import org.elasticsearch.threadpool.ThreadPool;
 import org.elasticsearch.xpack.core.inference.action.InferenceAction;
 import org.elasticsearch.xpack.esql.inference.InferenceRunner;
 import org.elasticsearch.xpack.esql.plugin.EsqlPlugin;
 
 import java.util.ArrayList;
 import java.util.List;
-import java.util.concurrent.ArrayBlockingQueue;
 import java.util.concurrent.BlockingQueue;
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Semaphore;
@@ -121,7 +121,7 @@ private ThrottledInferenceRunner(InferenceRunner inferenceRunner, ExecutorServic
             this.executorService = executorService;
             this.permits = new Semaphore(maxRunningTasks);
             this.inferenceRunner = inferenceRunner;
-            this.pendingRequestsQueue = new ArrayBlockingQueue<>(maxRunningTasks, true);
+            this.pendingRequestsQueue = ConcurrentCollections.newBlockingQueue();
         }
 
         public static ThrottledInferenceRunner create(