Ensuring LB fails with ListAgentPoolLBs: failed to get agent pool vmSet names

[CecileRobertMichon]

What happened:

Updated cloud-provider to v0.7.3 and CAPZ test E2E fail with this error:

I0419 22:08:22.703254       1 controller.go:368] Ensuring load balancer for service default/web-ilb
I0419 22:08:22.703385       1 azure_loadbalancer.go:1429] reconcileLoadBalancer for service(default/web-ilb) - wantLb(true): started
I0419 22:08:22.703631       1 event.go:291] "Event occurred" object="default/web-ilb" kind="Service" apiVersion="v1" type="Normal" reason="EnsuringLoadBalancer" message="Ensuring load balancer"
I0419 22:08:22.748645       1 azure_backoff.go:311] LoadBalancerClient.List(capz-e2e-7hbnku) success
E0419 22:08:22.748769       1 azure_loadbalancer.go:1433] reconcileLoadBalancer: failed to reconcile shared load balancer: %!w(*fmt.wrapError=&{reconcileSharedLoadBalancer: failed to list LB: ListAgentPoolLBs: failed to get agent pool vmSet names: not a vmss instance 0xc0006213c0})
E0419 22:08:22.748815       1 azure_loadbalancer.go:104] reconcileLoadBalancer(default/web-ilb) failed: reconcileSharedLoadBalancer: failed to list LB: ListAgentPoolLBs: failed to get agent pool vmSet names: not a vmss instance
E0419 22:08:22.748934       1 controller.go:275] error processing service default/web-ilb (will retry): failed to ensure load balancer: reconcileSharedLoadBalancer: failed to list LB: ListAgentPoolLBs: failed to get agent pool vmSet names: not a vmss instance
I0419 22:08:22.749407       1 event.go:291] "Event occurred" object="default/web-ilb" kind="Service" apiVersion="v1" type="Warning" reason="SyncLoadBalancerFailed" message="Error syncing load balancer: failed to ensure load balancer: reconcileSharedLoadBalancer: failed to list LB: ListAgentPoolLBs: failed to get agent pool vmSet names: not a vmss instance"

https://storage.googleapis.com/kubernetes-jenkins/pr-logs/pull/kubernetes-sigs_cluster-api-provider-azure/1323/pull-cluster-api-provider-azure-e2e-full/1384253528750428160/artifacts/clusters/capz-e2e-7hbnku/kube-system/cloud-controller-manager/cloud-controller-manager.log

This seems to be a regression from #588. This is on a cluster without the multiple SLBs feature.

What you expected to happen: Both VMAS and VMSS nodes should be supported for clusters in "vmss" vmType with Standard LB, as stated in #338 (comment).

How to reproduce it:

Anything else we need to know?:

Environment:

• Kubernetes version (use kubectl version):
• OS (e.g. from /etc/os-release):
• Kernel (e.g. uname -a):
• Install tools:
• Others: