applying HA settings may need to check for slow K8s API responses

### Current Behavior

We have seen a single occurrence of new NR instance failing to restart after having HA enabled.

The instance appears to have failed to create some of the underlying k8s resources because the k8s api was slow to remove them as part of the suspending the instance to apply the settings.

### Expected Behavior

Clean restart

### Steps To Reproduce

Not possible to reporduce as we do not know why the underlying k8s api was slow, the api is async but we do have some polling in place to check when k8s entities are removed.

### Environment

- FlowFuse version: 2.21.2
- Node.js version: 20.x
- npm version: 9.x
- Platform/OS: FFC
- Browser: ALL

### Have you provided an initial effort estimate for this issue?

I have provided an initial effort estimate