Backport of scheduler: volume updates should always be destructive into release/1.3.x

[hc-github-team-nomad-core]

Backport

This PR is auto-generated from #13008 to be assessed for backporting due to the inclusion of the label backport/1.3.x.

The below text is copied from the body of the original PR.

---

Fixes #12963

No task driver with file system isolation can support updating mounts in-place. Host volume and CSI volume or volume mount additions or updates should therefore always be destructive.

---

This changeset includes some unit testing but I've manually tested it on real CSI volumes as well. Adding a new volume:

$ nomad job plan ./demo/csi/hostpath/redis.nomad
+/- Job: "example"
+/- Task Group: "cache" (1 create/destroy update)
  +   Volume {
      + AccessMode:     "single-node-reader-only"
      + AttachmentMode: "file-system"
      + Name:           "volume0"
      + PerAlloc:       "true"
      + ReadOnly:       "true"
      + Source:         "test-volume"
      + Type:           "csi"
      }
  +/- Task: "redis" (forces in-place update)

Altering an existing volume:

$ nomad job plan ./demo/csi/hostpath/redis.nomad
+/- Job: "example"
+/- Task Group: "cache" (1 create/destroy update)
  +/- Volume {
    +/- AccessMode:     "single-node-reader-only" => "single-node-writer"
        AttachmentMode: "file-system"
        Name:           "volume0"
        PerAlloc:       "true"
    +/- ReadOnly:       "true" => "false"
        Source:         "test-volume"
        Type:           "csi"
      }
      Task: "redis"

[github-actions]

I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.