Fix a problem where it takes a long time to close a handle

Author: gmfeinberg

CHANGELOG.md

@@ -8,6 +8,12 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/).
 - Cloud only: added new OCI regions (TYO, AHU, DAC, DOH, IZQ)
 - Cloud only: added use of ETag in AddReplicaRequest and DropReplicaRequest
 
+### Fixed
+- Changed handle close to not use the Netty "quiet period" when shutting down
+  the worker group. This makes handle close much faster as well as making it
+  faster to acquire authentication information in the cloud service when using a
+  method that uses an HTTP request such as Instance Principal authentication
+
 ## [5.4.14] 2023-11-17
 
 ### Added

driver/src/main/java/oracle/nosql/driver/httpclient/HttpClient.java

@@ -366,7 +366,16 @@ ConnectionPool getConnectionPool() {
      */
     public void shutdown() {
         pool.close();
-        workerGroup.shutdownGracefully().syncUninterruptibly();
+        /*
+         * 0 means no quiet period, waiting for more tasks
+         * 5000ms is total time to wait for shutdown (should never take this
+         * long
+         *
+         * See doc:
+         * https://netty.io/4.1/api/io/netty/util/concurrent/EventExecutorGroup.html#shutdownGracefully--
+         */
+        workerGroup.shutdownGracefully(0, 5000, TimeUnit.MILLISECONDS).
+            syncUninterruptibly();
     }
 
     public Channel getChannel(int timeoutMs)