Drop kube-controller-manager workaround

[fdegir]

It's been identified that the use of EIB Kubernetes manifests mechanism to update the HelmChartConfigs for the core Helm charts results in race condition, causing the Kubernetes kube-controller-manager to throw an unhandled exception due to both RKE2 HelmController and kube-controller-manager to operate on the same resource.

To avoid triggering the issue, EIB documentation must be updated to instruct users not to use EIB Kubernetes manifests and instead use EIB os-files.

This PR is proposed to ensure the workaround can be dropped from main branch and backported to 3.3 for the upcoming maintenance release as use of os-files is successfully validated.

[fdegir]

@hardys @dbw7 I'm not sure if we already have the EIB documentation updated to instruct users about the use of EIB Kubernetes manifest mechanism for updating HelmChartConfigs causing the race condition. If not, it should perhaps be done. It can perhaps be explicitly stated in the release notes as well once EIB documentation is updated.

[hardys]

I discussed this with @dbw7 - I think we need to ensure the os-files workaround is actually documented somewhere before dropping the current workaround.

I guess this can be a general note related to configuration of core charts, and we definitely need to check here and https://github.com/suse-edge/atip to ensure any examples are handled correctly and are aligned with what we've found to work in SV.

[dbw7]

Should be good to do, just merged in main for EIB and backported to 1.1 and 1.2.

[hardys]

Pushed #802 which adjusts what I think is the only relevant example in the docs, so we can merge that, then this PR and backport both to 3.3