Internal: ClusterStateObserver should log on trace on timeout

This more consistent with the other logging it makes and since it can be used in many operations the output can be more verbose (without adding too much info as to who timed out exactly - which we can fix separately). If need be the caller of the observer can log a higher level message.

Closes #11722

core/src/main/java/org/elasticsearch/cluster/ClusterStateObserver.java

@@ -120,7 +120,7 @@ public class ClusterStateObserver {
                 timeoutTimeLeftMS = timeOutValue.millis() - timeSinceStartMS;
                 if (timeoutTimeLeftMS <= 0l) {
                     // things have timeout while we were busy -> notify
-                    logger.debug("observer timed out. notifying listener. timeout setting [{}], time since start [{}]", timeOutValue, new TimeValue(timeSinceStartMS));
+                    logger.trace("observer timed out. notifying listener. timeout setting [{}], time since start [{}]", timeOutValue, new TimeValue(timeSinceStartMS));
                     // update to latest, in case people want to retry
                     timedOut = true;
                     lastObservedState.set(new ObservedState(clusterService.state()));
@@ -232,7 +232,7 @@ public class ClusterStateObserver {
             if (context != null) {
                 clusterService.remove(this);
                 long timeSinceStartMS = TimeValue.nsecToMSec(System.nanoTime() - startTimeNS);
-                logger.debug("observer: timeout notification from cluster service. timeout setting [{}], time since start [{}]", timeOutValue, new TimeValue(timeSinceStartMS));
+                logger.trace("observer: timeout notification from cluster service. timeout setting [{}], time since start [{}]", timeOutValue, new TimeValue(timeSinceStartMS));
                 // update to latest, in case people want to retry
                 lastObservedState.set(new ObservedState(clusterService.state()));
                 timedOut = true;