[K8SPG-165] Optional Resource requests/limits for operator containers

[djtaylor]

Cleaned up stray commits from previous attempt at a PR. This provides a way to specify optional resource requests/limits for the containers in the operator. In a Kubernetes cluster where resource quotas are mandatory, these will always fail to start unless specified.

See: https://jira.percona.com/projects/K8SPG/issues/K8SPG-165?filter=allopenissues

[JNKPercona]

Test name	Status
upgrade	failed
demand-backup	failed
init-deploy	failed
scheduled-backup	failed
scaling	failed
smart-update	failed
recreate	failed
version-service	failed
affinity	failed
monitoring	failed
self-healing	failed
operator-self-healing	failed
clone-cluster	failed
tls-check	failed
users	failed
ns-mode	failed

commit: https://github.com/percona/percona-postgresql-operator/pull/193/commits/96e43a8fdaff4a2861102caa102ef443c460126d image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-pgo-apiserver

image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-pgo-event

image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-pgo-rmdata

image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-pgo-scheduler

image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-postgres-operator

image: perconalab/percona-postgresql-operator:PR-193-96e43a8f-pgo-deployer

[cap1984]

@djtaylor please follow @hors suggestions. Baseline looks good to me

[it-percona-cla]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}