[FLINK-5411] [flip6] Fix JobLeaderIdService shut down in ResourceManager

The JobLeaderIdService was formerly closed at two different locations. Once in the
ResourceManager and once in the ResourceManagerRuntimeServices. Since the JobLeaderIdService
is a RM specific component. It should also be closed in the scope of the RM.

This closes apache#5757.

flink-runtime/src/main/java/org/apache/flink/runtime/resourcemanager/ResourceManagerRunner.java

@@ -20,7 +20,6 @@
 
 import org.apache.flink.configuration.Configuration;
 import org.apache.flink.runtime.clusterframework.types.ResourceID;
-import org.apache.flink.runtime.concurrent.FutureUtils;
 import org.apache.flink.runtime.entrypoint.ClusterInformation;
 import org.apache.flink.runtime.heartbeat.HeartbeatServices;
 import org.apache.flink.runtime.highavailability.HighAvailabilityServices;
@@ -106,13 +105,7 @@ public CompletableFuture<Void> closeAsync() {
 		synchronized (lock) {
 			resourceManager.shutDown();
 
-			return FutureUtils.runAfterwards(
-				resourceManager.getTerminationFuture(),
-				() -> {
-					synchronized (lock) {
-						resourceManagerRuntimeServices.shutDown();
-					}
-				});
+			return resourceManager.getTerminationFuture();
 		}
 	}
 

flink-runtime/src/main/java/org/apache/flink/runtime/resourcemanager/ResourceManagerRuntimeServices.java

@@ -45,12 +45,6 @@ public JobLeaderIdService getJobLeaderIdService() {
 		return jobLeaderIdService;
 	}
 
-	// -------------------- Lifecycle methods -----------------------------------
-
-	public void shutDown() throws Exception {
-		jobLeaderIdService.stop();
-	}
-
 	// -------------------- Static methods --------------------------------------
 
 	public static ResourceManagerRuntimeServices fromConfiguration(