Pod group: Replacement Pods should be added as Workload owners

[alculquicondor]

What happened:

When a Pod fails, it can be replaced with a new pod. However, the new pod doesn't become an owner of the Workload.

This could lead to a Workload from disappearing before all the Pod finalizers have been removed.

What you expected to happen:

The replacement Pod should be owner of the Workload. Excess pods should not be owners.

How to reproduce it (as minimally and precisely as possible):

1. Create pod group
2. Fail one pod
3. Create replacement pod

Anything else we need to know?:

Environment:

• Kubernetes version (use kubectl version):
• Kueue version (use git describe --tags --dirty --always):
• Cloud provider or hardware configuration:
• OS (e.g: cat /etc/os-release):
• Kernel (e.g. uname -a):
• Install tools:
• Others:

[alculquicondor]

cc @achernevskii

[alculquicondor]

From @mimowo:

re-admitted workload gets deleted after completing. This is because the replacement pods don't become owners of the workload. So the workload gets deleted by garbage-collector very quickly

[mimowo]

/assign
I would like to give it a try, continuing tasks related to pod groups