[BACKPORT][v1.6.1][BUG] Executing fstrim while rebuilding causes IO errors

[github-actions]

backport #7103

[longhorn-io-github-bot]

Pre Ready-For-Testing Checklist

• Where is the reproduce steps/test steps documented?
  The reproduce steps are at: [BUG] Executing fstrim while rebuilding causes IO errors #7103 (comment).
  The test steps are at: Don't return an error on unmap while rebuilding longhorn-engine#1008 (comment).

• Is there a workaround for the issue? If so, where is it documented?
  Just wait for the rebuild to complete and attempt the trim again. However, VM workloads will remain paused for the duration of the rebuild.

• Does the PR include the explanation for the fix or the feature?

• Have the backend code been merged (Manager, Engine, Instance Manager, BackupStore etc) (including backport-needed/*)?
  The PR is at: Don't return an error on unmap while rebuilding (backport #1008) longhorn-engine#1011.

• If labeled: require/doc Has the necessary document PR submitted or merged (including backport-needed/*)?
  The documentation issue/PR is at: Explain trim behavior while rebuilding and clean up doc website#868

[yangchiu]

Verified passed on v1.6.x-head (longhorn-engine 1f6265f) following the test plan.