Move Kubernetes 410 exception handling outside generator

[kevingrismore]

The previous implementation tried to catch 410 exceptions inside the _job_events generator, but continued reports of 410s indicate that it's likely that strategy would only detect a 410 if raised on the initial call to stream. This moves handling and retrying the watch to the caller.

Checklist

• This pull request references any related issue by including "closes <link to issue>"
  • If no issue exists and your change is not a small fix, please create an issue first.
• If this pull request adds new functionality, it includes unit tests that cover the changes
• If this pull request removes docs files, it includes redirect settings in mint.json.
• If this pull request adds functions or classes, it includes helpful docstrings.

[kevingrismore]

Hoping to get a check on whether this change is sensible and actually makes a difference. The original test still passes since the watch is being restarted, but I'm wondering if there's a better way to test this that proves moving the handling up a layer matters, and when the exception is raised by the stream has a real impact.