Bug 1999571

Description of problem:
When cloning an nfs DV with a target size that is bigger than the source, clone does not progress

Version-Release number of selected component (if applicable):
CNV 4.9.0

How reproducible:
100%

Steps to Reproduce:
1. Clone nfs DV (manifests below)

Actual results:
NAMESPACE                            NAME        PHASE       PROGRESS   RESTARTS   AGE
default                              dv-target                                     3s

"level":"error","ts":1630407986.6950176,"logger":"controller-runtime.manager.controller.datavolume-controller","msg":"Reconciler error","name":"dv-target","namespace":"default","error":"source/target sizes not compatible"

Expected results:
DV Succeeded

Additional info:
- In 4.8 this operation succeeds
- Are we ok with looping over https://github.com/kubevirt/containerized-data-importer/blob/main/pkg/controller/datavolume-controller.go#L1558-L1560 indefinitely (it's part of advancedClonePossible check)? shouldn't we just fail here and go to host assisted?
- The only indication of error is in cdi-deployment logs
- When reproducing, keep in mind that the nfs pvc.Status.Capacity can be bigger than your request in the DV

[cnv-qe-jenkins@alex490-143-szbxl-executor clone-nfs-loop]$ cat dv_source.yaml 
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata:
  name: dv-source
  namespace: openshift-virtualization-os-images
spec:
  source:
      http:
         url: "http://.../rhel-84.qcow2"
  pvc:
    storageClassName: nfs
    namespace: default
    accessModes:
    - ReadWriteOnce
    resources:
      requests:
        storage: 25Gi
[cnv-qe-jenkins@alex490-143-szbxl-executor clone-nfs-loop]$ cat dv_target.yaml 
apiVersion: cdi.kubevirt.io/v1beta1
kind: DataVolume
metadata: 
  name: dv-target
  namespace: default
spec: 
  pvc:
    storageClassName: nfs
    accessModes: 
      - ReadWriteOnce
    resources: 
      requests: 
        storage: 26Gi
  source: 
    pvc: 
      name: dv-source
      namespace: openshift-virtualization-os-images