Overlapping NodeConfigs don't track their resources correctly and fight with each other

[mflendrich]

Issue originally authored by tnozicka as #2051

When you create 2 NodeConfigs selecting the same Kubernetes Node but doing different things like "tuning" vs "disk setup", e.g. the "tuning" one with remove the other's mount units.

I0802 19:44:23.525788       1 systemd/unit.go:132] "Disabling and stopping unit because it's no longer required" Name="mnt-persistent\\x2dvolumes.mount"
I0802 19:44:23.544288       1 systemd/unit.go:142] "Removing unit because it's no longer required" Name="mnt-persistent\\x2dvolumes.mount"
I0802 19:44:23.545521       1 record/event.go:376] "Event occurred" object="e2e-cluster" fieldPath="" kind="NodeConfig" apiVersion="scylla.scylladb.com/v1alpha1" type="Normal" reason="MountDeleted" message="Mount unit mnt-persistent\\x2dvolumes.mount has been deleted"