Update CAPI support to 1.9

[Danil-Grigorev]

This is an epic for 1.9 update, and changes required to facilitate that.

• Update CAPI version to 1.9.x
• Evaluate issues with docker templates #530
• Pre etcd membership change draining and volume mount detachment handling issues - Machine fails to finish draining/volume detachment after successful completion kubernetes-sigs/cluster-api#11591

[Danil-Grigorev]

Issues can be reproduced on #526 which only updates CAPI to 1.9 without changing the code. It appears kubernetes-sigs/cluster-api#11457 has a check which resolves the problem, but it applies only for kubeadm CP provider.

[Danil-Grigorev]

After some investigation opened kubernetes-sigs/cluster-api#11591 and a fix kubernetes-sigs/cluster-api#11590 which unblocked machine deletion after update

[Danil-Grigorev]

From the discussion it seems there is https://github.com/kubernetes-sigs/cluster-api/blob/main/docs/proposals/20240916-improve-status-in-CAPI-resources.md#machine-newconditions proposes removal of existing conditions in the future, which acts as a way to ensure draining success and volume detachment success does terminate further attempts.