Finish health check polling in status monitor

[AustEcon]

complete current model as per #57

[AustEcon]

Actually this model will not work for components that do not have a network API... so the correct model would be more like a supervisor wrapping each processes that can check the returncode upon exit (or when killed by some other method)... which would then update the component_state.json file and in turn the status_monitor would notice the change and send a push notification reflecting the change.