Add worker health output to __heartbeat__ endpoint.

[Micheletto]

The __heartbeat__ endpoint should be authoritative for the health of the service. If backend services, or workers are not functioning, the endpoint should reflect that with error status and output. A json list of backend, and their present status is used in other services and would be helpful here but not strictly necessary.

[g-k]

Since the scanners shut themselves down, I think we can add an hourly or test scan to the runner / cron process then treat zero scansLast24Hours from /api/v1/stats?format=text as a backend being down.