feat: wire TrainJobStatus server into controller manager readyz and healthz probes

[krishdef7]

What you would like to be added?

The TrainJobStatus server introduced in #3227 runs in the same process as the controller manager but is not registered with its /readyz and /healthz probes.

The status server should implement healthz.Checker and register via mgr.AddHealthzCheck and mgr.AddReadyzCheck in pkg/statusserver/setup.go.

Discussed in #3227 and flagged as a follow-up by @andreyvelich.

/kind feature
/area controller

Why is this needed?

Training pods have no way to verify the status server is ready before sending their first update, causing silent failures if the server is still initializing TLS or the OIDC provider. If the server crashes mid-job, the controller pod stays Running with no signal from the existing liveness probe.

Wiring into readyz/healthz is a low-risk pattern already used by the webhook server in the same process.

Love this feature?

Give it a 👍 We prioritize the features with most 👍

[github-actions]

🎉 Welcome to the Kubeflow Trainer! 🎉

Thanks for opening your first issue! We're happy to have you as part of our community 🚀

Here's what happens next:

• Our team will review your issue soon! cc @kubeflow/kubeflow-trainer-team
• If you'd like to contribute to this issue, check out our Contributing Guide for repo-specific guidelines and the Kubeflow Contributor Guide for general community standards.

Join the community:

• Slack: Join our #kubeflow-trainer Slack channel.
• Meetings: Attend the Kubeflow AutoML and Training Working Group bi-weekly meetings.

Feel free to ask questions in the comments if you need any help or clarification!
Thanks again for contributing to Kubeflow! 🙏

[krishdef7]

/assign