Fix: Ensure ConfigMap and StatefulSet updates are applied during operator upgrades

[Mangaal]

What type of PR is this?
This PR fixes an issue where updates to the argocd-redis-ha-configmap, argocd-redis-ha-health-configmap, and the StatefulSet argocd-redis-ha-server were not being reflected during OpenShift GitOps operator upgrades.

The current code missed checking for changes in:

• The data field of argocd-redis-ha-configmap and argocd-redis-ha-health-configmap.
• The env variables in the StatefulSet argocd-redis-ha-server.

As a result, changes introduced in higher versions of the operator were not applied unless the config maps and StatefulSet were manually deleted or the .spec.ha.enabled field was toggled in the ArgoCD Custom Resource (CR). This PR ensures these changes are correctly detected and applied.

This PR:
Adds logic to check for changes in the argocd-redis-ha-configmap and argocd-redis-ha-health-configmap data fields during operator upgrades.
Includes validation for the env variables in the argocd-redis-ha-server StatefulSet to ensure updates are reflected.
Ensures that any required updates are applied automatically without manual intervention.

/kind bug

Steps to reproduce the issue:

• Install OpenShift GitOps Operator v1.12.0 on an OCP 4.16.4 cluster.
• Create an ArgoCD instance with .spec.ha.enabled = true.
• Upgrade the operator to v1.14.0, v1.14.1, and v1.14.2.
• Observe that the argocd-redis-ha-configmap and argocd-redis-ha-health-configmap are not updated as expected.
• Delete the config maps manually, and observe that they are recreated correctly but result in AUTH errors.
• Toggle the .spec.ha.enabled field to fix the issue.

[svghadi]

Can we add a unit test to check configmap reconciliation? Create the cm resource using reconcileRedisHAHealthConfigMap function, modify the created configmap resource and run the reconcileRedisHAHealthConfigMap again to ensure the values are reset to default

[svghadi]

Just a small nit, everything else looks great! Awesome job on the unit tests. Thanks!

[svghadi]

Hi, Sorry nitpicking again. Your previous comments made sense in 2 places. Can you revert the changes for this. Also please rebase, I have just merged a PR which will fix failing extra-cmd e2e test.

[svghadi]

--- FAIL: kuttl/harness/1-036_validate_role_rolebinding_for_source_namespace (64.05s)

is a known flaky failure. Ignoring it and merging the PR.

[svghadi]

/cherry-pick release-0.13

[gcp-cherry-pick-bot]

Cherry-pick failed with Merge error 5049cdfee963972964cfb89d510274136802c3d5 into temp-cherry-pick-32de84-release-0.13