Execution status compute API

[wdbaruni]

Expose an API (based on NATS) in the compute node to allow the requester node to query about the status and progress of jobs. This should be similar to the periodic messages the compute node is publishing #3859, but enables the requester to query about certain jobs that it might've missed hearing about, such as when a compute node says jobA is not among the active ones, but the requester's state says otherwise.

This may not be needed if we implement #3859 in a way using offsets that makes sure on messages were missed and for the requester to go back to previous events

[wdbaruni]

Not needed based on the design proposed for #3861 where we maintain offsets of latests processed events, along with periodic state sync