Resilient Workers

[unexge]

Currently we don't have any way to detect or recover from Worker failures. We should be able to detect if a Worker is failed (we might check status of the process or use some kind of health checking) and if that's the case, we should either indicate that Worker is failed to the Pool and then the Pool can replace the Worker with a new one or we can recover from the failure silently in the Worker.

Relevant modules:
Worker: https://github.com/masomo/coyote/blob/main/src/worker/worker.rs
Pool: https://github.com/masomo/coyote/blob/main/src/worker/pool.rs, https://github.com/masomo/coyote/blob/main/src/worker/pool/static_.rs