Support node selector for each profile

[songtao98]

Is your feature request related to a problem? Please describe.

There are use cases such as to config HighNodeUtilization plugin with different threshold in different group of nodes, e.g. nodepools. However, most of the plugins only supports to filter pods through specific filters.

Describe the solution you'd like

When dealing with nodes, it is the framework to decide which nodes can be the targets. And it is configured by a global NodeSelector. So the idea is to add a new arg into Profile. Nodes that passed the global NodeSelector will be filtered again by each profile's NodeSelector and then passed into pg.Balance() or pg.Deschedule(). However, this might change the way how you build plugins.

Describe alternatives you've considered

For each plugin, add NodeSelector configuration to filter nodes passed into it.

What version of descheduler are you using?

descheduler version: v0.26.0 and later

Additional context

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale