[BACKPORT][v1.5.4][IMPROVEMENT] Remove startup probe of CSI driver after liveness probe conn fix ready

[github-actions]

backport #7428

[ejweber]

I know it is late in the v1.5.4 cycle, but @PhanLe1010, @james-munson, and I discussed it and agree that it is better to get these changes merged before v1.5.4 releases. It is because no version of v1.5.x has gone out yet with the workaround fix I implemented. This final fix is safer and better.

[longhorn-io-github-bot]

Pre Ready-For-Testing Checklist

• Where is the reproduce steps/test steps documented?
  The test steps are at: Revert "Add a startup probe to the longhorn-csi-plugin container" longhorn-manager#2578 (comment).

• Is there a workaround for the issue? If so, where is it documented?
  No workaround is needed. This is an improved fix for an issue we already temporarily fixed.

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

• Does the PR include deployment change (YAML/Chart)? If so, where are the PRs for both YAML file and Chart?
  The PR for the YAML change is at: Use livenessprobe version with upstream fix (backport #7908) #7935.
  The PR for the chart change is at: Use livenessprobe version with upstream fix (backport #7908) #7935.

• Have the backend code been merged (Manager, Engine, Instance Manager, BackupStore etc) (including backport-needed/*)?
  The PR is at: Revert "Add a startup probe to the longhorn-csi-plugin container" (backport #2578) longhorn-manager#2583.

[chriscchien]

Verified pass on longhorn v1.5.x (longhorn-manager b2410e) with test steps

Change service longhorn-backend to port 5001 then delete any longhorn-csi-plug-xx pod, after pod recreated, container longhorn-csi-plugin and node-driver-registrar will keep in state CrashLoopBackOff, container longhorn-liveness-probe will keep in running state.
Change service longhorn-backend back to the original port, all 3 containers in pod longhorn-csi-plug-xx become in running state.