Backport #8421 to v1.5: flush graceful Disassociate PDUs before transport force-close on shutdown

ActorTransportAdapter.Shutdown() started manager.GracefulStop() and
WrappedTransport.Shutdown() concurrently. The manager's child association actors
(the ProtocolStateActors under AkkaProtocolTransport) write the graceful
Disassociate PDU as part of their own shutdown, so running the wrapped-transport
teardown concurrently let the underlying transport force-close the channel before
that PDU reached the wire. Peers were then left with a zombie association until the
transport failure detector tripped (acceptable-heartbeat-pause, 120s by default);
under the trttl/gremlin test adapters, which swallow the raw TCP close, that lost
PDU is the sole prompt teardown signal.

Sequence the two operations: stop the manager first (bounded by flush-wait-on-
shutdown), then tear down the wrapped transport. The wrapped transport is torn
down whether or not the manager stopped cleanly, so shutdown cannot hang on this
ordering.

(cherry picked from commit a918393)

Author: Aaronontheweb

src/core/Akka.Remote/Transport/TransportAdapters.cs

@@ -550,8 +550,21 @@ protected override void InterceptAssociate(Address remoteAddress, TaskCompletion
         public override Task<bool> Shutdown()
         {
             var stopTask = manager.GracefulStop((RARP.For(System).Provider).RemoteSettings.FlushWait);
-            var transportStopTask = WrappedTransport.Shutdown();
-            return Task.WhenAll(stopTask, transportStopTask).ContinueWith(x => x.IsCompleted && !(x.IsFaulted || x.IsCanceled), TaskContinuationOptions.ExecuteSynchronously);
+
+            // Sequence the wrapped-transport teardown AFTER the manager (and its child association
+            // actors) have stopped. Those children write the graceful Disassociate PDU as part of
+            // their own shutdown; running WrappedTransport.Shutdown() concurrently (as this used to)
+            // let the underlying transport force-close the channel out from under a not-yet-flushed
+            // Disassociate. Peers then kept a "zombie" association until the transport failure
+            // detector tripped (up to acceptable-heartbeat-pause, 120s by default). The manager stop
+            // is bounded by flush-wait-on-shutdown, and the wrapped transport is torn down regardless
+            // of whether the manager stopped cleanly, so shutdown can never hang on this ordering.
+            var transportStopTask = stopTask.ContinueWith(_ => WrappedTransport.Shutdown(),
+                TaskContinuationOptions.ExecuteSynchronously).Unwrap();
+
+            return Task.WhenAll(stopTask, transportStopTask)
+                .ContinueWith(x => x.IsCompleted && !(x.IsFaulted || x.IsCanceled),
+                    TaskContinuationOptions.ExecuteSynchronously);
         }
     }