🐛 [BUG] - Hibernated clusters should not be onboarded

[IvoGoman]

Priority

None

Description

I have hibernated a Gardener shoot and removed the cluster from Greenhouse. It is still labeled matching a CareInstruction.
After some time the shoot is onboarded again to greenhouse, which also results in Plugins being created. Neither the cluster or the plugins will be ready in this state. This results in prometheus alerts for the involved components.

My expectation was that a hibernated cluster should not be touched by shoot grafter, as it is not currently up-and-running.

Reproduction steps

1. Go to Gardener Dashboard and hibernate a cluster
2. Label the cluster to match the CareInstruction
3. See the cluster being onboarded

Manifests

apiVersion: greenhouse.sap/v1alpha1
  kind: ...

Screenshots

![DESCRIPTION](LINK.png)

[uwe-mayer]

Should be solved with the correct CEL expression targeting shoots.
E.g.

shootSelector:
    expression: has(object.status.conditions) && object.status.conditions.exists(c,
      c.type == "SystemComponentsHealthy" && c.status == "True")