Add heartbeat process to detect thread pool exhaustion

#### What problem would the feature you're requesting solve? Please describe.
Currently, there is no automated diagnostic in place that would detect deadlock/thread pool exhaustion in the Host. The only mode we have right now to detect thread pool exhaustion is by manually capturing and examining a memory dump.

#### Describe the solution you'd like
Add a heartbeat monitor process that pings the host on a regular schedule, and detects thread pool exhaustion scenarios.

#### Additional context
Once the heartbeat is in place, we will add one of two options for mitigation of the condition detected:

1. Auto-restart the host
or
2. Report this information through the status API's to the Antares platform for auto-mitigation/ memory dump collection.

No files or tests are named. Begin by identifying the Host entry point and its existing status API, then determine how a scheduled heartbeat should detect thread-pool exhaustion. Done should include a defined detection signal and an agreed mitigation path: host restart or reporting the condition through the status APIs.

Written by the indexing model from the issue text.