Support for "safe-to-evict" annotation

[Radoslaw-Kutylowski]

Describe the feature
Kubernetes autoscalers use a "safe-to-evict" annotation (https://kubernetes.io/docs/reference/labels-annotations-taints/#cluster-autoscaler-kubernetes-io-safe-to-evict) to identify pods that should not be evicted while they are running. Would it be possible to add support for this, or a similar solution, in NTH? In many cases it would probably need a grace period after which a forceful drain would occur (e.g. in case of spot disruption).

Is the feature request related to a problem?
We have Kubernetes Jobs that are running tests. Sometimes these jobs are terminated e.g. by rebalance recommendations, that could wait a short while for the job to finish. This disrupts our CD workflows.

Describe alternatives you've considered
Outside of being open to other labels/annotations being used instead of the default "cluster-autoscaler" one, I do not have alternative ideas

[youwalther65]

Another use case I came across is long-running Kubernetes Jobs and ASG instance refresh. It would be great to wait for job to finish here.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you want this issue to never become stale, please ask a maintainer to apply the "stalebot-ignore" label.

[github-actions]

This issue was closed because it has become stale with no activity.