worker.py: Avoid letting a job get stuck if exception occurs

[JonathonHall-Purism]

Some instances of stuck jobs were observed recently for PureOS. From the logs, I think a Python exception may have occurred after the build completed but before the artifacts were uploaded. I can't tell what might have caused that exception, if it did occur.

This change would ensure that a 'rejected' status is sent if this occurs, rather than leaving the job stuck with no result.

I applied this change to fennel.pureos.net (the new worker) to try to identify what was causing the stuck jobs, but this never happened again. No jobs got stuck after that and this exception code was never hit, so I can't identify the root cause.

[JonathonHall-Purism]

I have not properly tested this change - I don't have a test environment set up, and this never happened again after I applied the change to fennel.pureos.net to see what was causing these jobs to get stuck this way.

The jobs were very old (there had been no amd64 worker for months), so it is possible that this would not happen if the job queue is being processed regularly.

So with that in mind, although it seems like a good robustness improvement to avoid leaving a job stuck if this happens, I understand if you'd rather reject this and see if it ever happens again or can be reproduced.