fix(statefulset): don't skip Delete when scale-to-0 Update fails

[dashashutosh80]

doDeleteStatefulSet returned early when the courtesy scale-to-0 Update failed (e.g. 409 Conflict from concurrent kube-controller-manager writes to .status), never reaching the actual Delete. recreateStatefulSet then silently discarded the error and called createStatefulSet, which observed the still-existing STS and failed with AlreadyExists. The operator finalised the reconcile as Completed while the host stayed at Replicas=0 with no further self-correction.

doDeleteStatefulSet:

• Treat scale-to-0 as best-effort. On Update failure, log a warning and proceed to Delete which terminates pods anyway.
• Treat IsNotFound on Get as success (contract: STS is gone).
• Skip scale-down when Replicas is already 0.
• Propagate Delete errors instead of swallowing them.

recreateStatefulSet:

• Stop discarding the error from doDeleteStatefulSet. Return a plain error so the next reconcile retries; ErrCRUDAbort would mark the whole CHI as ReconcileAbort and require manual intervention, which is too aggressive for a transient conflict.

Important items to consider before making a Pull Request

Please check items PR complies to:

• All commits in the PR are squashed. More info
• The PR is made into dedicated next-release branch, not into master branch¹. More info
• The PR is signed. More info

This issue fixes #1990