Implement Pod Failure Policy to react early on transient failures

[devantler]

Description

Currently, KSail does not fail early in case a release is failing. It will keep retrying until an arbitrary timeout has been reached. Ideally it should fail immediately as it encounters a non-retriable Pod failure.

This issue suggests using the new Pod Failure Policy to alleviate this:

• Kubernetes 1.31: Pod Failure Policy for Jobs

Current behavior

• KSail only fails if an arbitrary timeout has been reached.

Proposed behavior

• KSail fails immediately as a non-retriable pod failure is identified.

Closes

No response