ssh-proxy-public service label fails to match a pod due to incorrect label and pod selector.

[sourix]

Describe the bug
In the KubeCF 2.5.8 release, when using LoadBalancers for the services (NOT ingress controller), it was noted that cf ssh did not work out of the box.
The scheduler pod has a label app.kubernetes.io/component=scheduler
The pod selector in the service definition for ssh-proxy-public looks for app.kubernetes.io/component=ssh-proxy
So quite obviously the pod selector in the service does not match the scheduler pod which has the ssh-proxy container.
The fix for this is quite simple and described in the additional context below and I'd be happy to submit a pull request if needed.

To Reproduce
Deploy KubeCF 2.5.8 on a Kubernetes backbone such as AWS EKS without nginx ingress controller.

Expected behavior
cf ssh should work.

Environment

• KubeCF 2.5.8
• ssh-proxy-public service
• scheduler pod.

Additional context
I edited the service ssh-proxy-public to look like below and can confirm that ssh works after that ...
Name: ssh-proxy-public Namespace: kubecf Labels: app.kubernetes.io/component=scheduler app.kubernetes.io/instance=kubecf app.kubernetes.io/managed-by=Helm app.kubernetes.io/name=kubecf app.kubernetes.io/version=2.5.8 helm.sh/chart=kubecf-2.5.8 Annotations: meta.helm.sh/release-name: kubecf meta.helm.sh/release-namespace: kubecf Selector: app.kubernetes.io/component=scheduler,app.kubernetes.io/instance=kubecf Type: LoadBalancer IP: XX.XX.XX.XX LoadBalancer Ingress: XX.XX.XX.XX Port: ssh 2222/TCP TargetPort: 2222/TCP NodePort: ssh 30017/TCP Endpoints: xx.xx.xx.xx:2222 Session Affinity: None External Traffic Policy: Cluster Events: <none>