Job Queue: improve logging / health monitoring

[tyteen4a03]

Currently, the Job Queue Worker doesn't emit a lot of metrics on its health, so it's hard to see first-hand whether the worker died or if it is just idling.

It would be great to be able to customize the logging options for the worker, as well as perhaps an endpoint /health that is only turned on for the worker for healthchecks.