Requesting for a feature much needed for debugging of failed jobs running on dynamic agents using Kubernetes plugin.

We want to persist the pods in case of job failures which will help team to debug the failure and can be retained for x no. of hours or days.