Restore fails with error "no space left on device" if PVCs are 100% utilised during backup

[nrb]

What steps did you take and what happened:
From an internal bug report:

Using Restic, restore is failing with error "no space left on device" if PVC was fully occupied by the user data while the backup is taken.

for example, PVC of size 1Gi and user data in the PVC is ~upto 100% which is of 1Gi

What did you expect to happen:

Velero restic restores should be successful, even if the PV is full.

The output of the following commands will help us better understand what's going on:
(Pasting long output into a GitHub gist or other pastebin is fine.)

root@k8-master-277:~# velero restore describe restore001 --details
Name:         restore001
Namespace:    velero
Labels:       <none>
Annotations:  <none>

Phase:  PartiallyFailed (run 'velero restore logs restore001' for more information)

Errors:
  Velero:   pod volume restore failed: error restoring volume: error creating .velero directory for done file: mkdir /host_pods/60880bf2-9d1c-47cf-ba8f-5f8db43b385b/volumes/kubernetes.io~csi/pvc-ee79e0f0-ed62-44f7-987e-16efca1d1cd5/mount/.velero: no space left on device
  Cluster:    <none>
  Namespaces: <none>

Anything else you would like to add:

Since this is related to creating the .velero donefile, the fix might be somewhat relate to #2722.

Environment:

• Velero version (use velero version): v1.4.0
• Cloud provider or hardware configuration: vSphere

Vote on this issue!

This is an invitation to the Velero community to vote on issues, you can see the project's top voted issues listed here.
Use the "reaction smiley face" up to the right of this comment to vote.

• 👍 for "I would like to see this bug fixed as soon as possible"
• 👎 for "There are more important bugs to focus on right now"

[nrb]

Scoping this for v1.6 currently; I'm unsure if the fix will be backport-able at this time.