[test] `testing.go:1712: race detected during execution of test`

### Which jobs are flaking?

[`periodic-cluster-api-test-mink8s-main`](https://prow.k8s.io/view/gs/kubernetes-ci-logs/logs/periodic-cluster-api-test-mink8s-main/2097550839932522496)

### Which tests are flaking?

```
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_DebugHandler
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_unstructured_CRUD
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_corev1_CRUD
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_rbacv1_CRUD
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_corev1_Watch_With_Cache
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_corev1_Watch
sigs.k8s.io/cluster-api/test/infrastructure/inmemory/pkg/server: TestAPI_PortForward
```

### Since when has it been flaking?

Only flaked once on `Sep 09 10:31:10 UTC+5:30`.

### Testgrid link

https://testgrid.k8s.io/cluster-api-core-main#capi-test-mink8s-main

### Reason for failure (if possible)

`testing.go:1712: race detected during execution of test`

### Anything else we need to know?

CI Signal team suspects this race to be quite a rare occurrence (job runs have consistently been green historically), but worth flagging nonetheless.

### Label(s) to be applied

/kind flake
/area testing

Start with the listed tests under test/infrastructure/inmemory/pkg/server and reproduce the periodic-cluster-api-test-mink8s-main failure, focusing on the race reported at testing.go:1712. Compare the affected TestAPI cases and use the linked CI log and Testgrid entry to investigate. Done means the race is understood and the affected tests pass reliably under the relevant job.

Written by the indexing model from the issue text.