Add graceful shutdown to worker instances

[SchahinRohani]

Description

This PR implements graceful shutdown for worker instances to ensure that the scheduler is properly notified when workers are shutting down.

How Has This Been Tested?

bazel test ...

Checklist

• Updated documentation if needed
• Tests added/amended
• bazel test //... passes locally
• PR is contained in a single commit, using git amend see some docs

---

This change is 

[adam-singer]

Reviewed 3 of 6 files at r1, all commit messages.
Reviewable status: 0 of 1 LGTMs obtained, and 3 of 6 files reviewed, and 3 discussions need to be resolved

---

nativelink-worker/src/local_worker.rs line 486 at r1 (raw file):

    pub async fn shutdown(&self) {
        println!("Shutting down worker: {}", self.name());

should be a tracing log statement and not println statements, for example

nativelink/nativelink-util/src/common.rs

Line 237 in 553f33c

event!(

---

nativelink-worker/src/local_worker.rs line 488 at r1 (raw file):

        println!("Shutting down worker: {}", self.name());

        let max_wait_duration = Duration::from_secs(60);

I think this would be better named as a shutdown deadline and a configuration option. I wonder if this would get out of sync with request timeouts and cause hard to catch behavior if shorter than upstream timeouts.

---

nativelink-worker/src/local_worker.rs line 525 at r1 (raw file):

            } else {
                println!("No worker_id available; cannot notify scheduler.");
                return;

This would be panic or return Err / make_err type situation