Add missing pvc namespace to the vrg rdspec

[nirs]

It seems that recent change for supporting multiple namespaces have broken failover with volsync when using multiple volsync apps. Add the pvc namespace to the protected pvc, similar to volrep code.

I did not modify the test to validate this change since I don't fully understand the test. It seems that there is a missing It section for testing the vrg.

Tested on ocp 4.14.

Issues:

• During upgrade on a system with stuck workload, the vrg is not updated automatically yet.

[nirs]

Example failure before this change

After failover the application is stuck in WaitingForResourceRestore

$ oc get drpc -A -o wide
NAMESPACE     NAME                           AGE   PREFERREDCLUSTER   FAILOVERCLUSTER   DESIREDSTATE   CURRENTSTATE   PROGRESSION                 START TIME             DURATION          PEER READY
test-cephfs   test-cephfs-placement-1-drpc   11m   cluster1           cluster2          Failover       FailingOver    WaitingForResourceRestore   2023-11-29T19:50:24Z                     False
test-rbd      test-rbd-placement-1-drpc      15m   cluster1           cluster2          Failover       FailedOver     Completed                   2023-11-29T19:42:43Z   3m11.406944081s   True

State of the failover cluster:

 oc get deploy,pod,pvc,vrg,vr,replicationsource,replicationdestination -n test-cephfs -o wide --context perf2
NAME                                STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS                AGE   VOLUMEMODE
persistentvolumeclaim/busybox-pvc   Bound    pvc-2456bbbf-c89c-4b64-8efa-65838dff33c8   1Gi        RWX            ocs-storagecluster-cephfs   11m   Filesystem

NAME                                                                       DESIREDSTATE   CURRENTSTATE
volumereplicationgroup.ramendr.openshift.io/test-cephfs-placement-1-drpc   primary        Secondary

NAME                                                 LAST SYNC              DURATION         NEXT SYNC
replicationdestination.volsync.backube/busybox-pvc   2023-11-29T19:50:21Z   1m2.651263526s   

In ramen-dr-cluster-operator log we see:

2023-11-29T19:54:26.257Z        INFO    controllers.VolumeReplicationGroup.vrginstance  controllers/volumereplicationgroup_controller.go:1012   Failed to restore PVs   {"VolumeReplicationGroup": "test-cephfs/test-cephfs-placement-1-drpc", "rid": "0e93bd0b-1851-4263-937b-11754deb2cfc", "State": "primary", "error": "failed to restore PVs for VolSync (failed to restore all PVCs using RDSpec ([{{ busybox-pvc true { { []} { []}} 0xc004482540 map[] map[app:test-cephfs [app.kubernetes.io/part-of:test-cephfs](http://app.kubernetes.io/part-of:test-cephfs) appname:busybox [apps.open-cluster-management.io/reconcile-rate:medium](http://apps.open-cluster-management.io/reconcile-rate:medium)] [ReadWriteMany] {map[] map[storage:{{1073741824 0} {<nil>} 1Gi BinarySI}] []} [] <nil> nil <nil>}}]))"}