VolumeAttributesClass update should support recover from failure

[xing-yang]

I opened an issue in k/k to track this: kubernetes/kubernetes#130597
The fix will probably mainly happen in csi-resizer.

[xing-yang]

cc @sunnylovestiramisu @gnufied @AndrewSirenko @jsafrane

[carlory]

related to: #483

[sunnylovestiramisu]

The suggested fix is:

When spec.vac == status.curvac but targetvac != status.curvac and the volume has a condition of isInfeasibleError(InvalidArgument, OutOfRange and NotFound), trigger modify volume(rollback) again and set targetvac=status.curvac, clear errors and conditions.

If the rollback operation fails, it will go back to the logic of "spec.vac == status.curvac but targetvac != status.curvac" as above, and retry modify volume with different rate based on the if the error is infeasible or final.

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[carlory]

/remove-lifecycle stale

[sunnylovestiramisu]

The fix is in #513