[Task] Split PVCs

[andrewazores]

          Could we extract this to a helper method and replace all 3 instances?

On that note, I suppose we can't get away with one PVC since these pods may be scheduled on other nodes and we default to the ReadWriteOnce mode. Maybe we could force all the pods to be on the same node using PodAffinity? We would need to merge this with whatever config the user provides in Scheduling Options, and that could get a bit tricky.

Assuming we stick with 3 PVCs, do you think it would make sense to have separate configurations for each PVC? I could see object storage requiring quite a bit larger size for recordings than the Postgres database. What is the main Cryostat pod's PVC used for now?

Originally posted by @ebaron in #923 (comment)

[andrewazores]

See also cryostatio/cryostat-helm#224