/kind bug

What steps did you take and what happened:
After upgrading CAPH, new control pane nodes are not added as targets to the API load balancer anymore.

What did you expect to happen:
No change in behaviour in this regard.

Anything else you would like to add:
Some control plane providers, at least the Talos one don't set this condition. According to https://github.com/kubernetes-sigs/cluster-api/blob/main/docs/proposals/20200506-conditions.md I understand this would be compliant behaviour:

A set of additional Machine.Status.Conditions should be created in order to provide an observation point on the operational state of kubeadm control-plane machine components like e.g. EtcdReady, ApiServerPodReady, etc. (the other control plane providers are expected to implement different conditions at this level)

While I do consider #1150 to be a great improvement for the consistency of the cluster, the underlying assumption that this condition will be present once the apiserver is ready seems to go too far. Thus, this feature should be optional and/or configurable to use e.g. request probes instead.

Environment:

• cluster-api-provider-hetzner version: v1.0.0
• Kubernetes version: v1.31.1
• OS (e.g. from /etc/os-release): Talos Linux v1.8.1