Now Monitoring driver is getting notifier instance to be used
in case of failure get data for a certain metric or failed to find
the metric added on a certain host, the notifier will be used to
notify admins list (notify-list in config file) that something is
wrong with the metrics and need to look at it
Base notification driver changed to support the following methods:
1. notify_status: will be used to notify if node is failed and
evacuated
2. notify: will be used anywhere in monitoring drivers to notify
administrators if something is wrong.
Change-Id: I12eddf03d1921a04f5fa9a8101a471bea5f9c507
61558d2b33