Improve logging around ansible timeouts

When a job gets killed by the timeout watchdog the log only contains
the note 'Ansible timeout exceeded' but no information about what the
timeout actually was. Add this to this message so unexpected timeouts
are easier to debug.

Change-Id: I4f7f29e649cf042e07408ec05cb4d938a209dddd

zuul/executor/server.py

@@ -1854,7 +1854,7 @@ class AnsibleJob(object):
         ret = None
         if timeout:
             watchdog = Watchdog(timeout, self._ansibleTimeout,
-                                ("Ansible timeout exceeded",))
+                                ("Ansible timeout exceeded: %s" % timeout,))
             watchdog.start()
         try:
             # Use manual idx instead of enumerate so that RESULT lines