Description of problem:
A cluster has been observed in the following state:
- It has a Machine which has been in a Deleting state for several weeks.
- The Node corresponding to the Machine has been successfully deleted.
- The AWS instance associated with the Machine no longer exists.

The machine-api-controllers just repeat that they cannot find the instance:

W1005 23:05:28.413511       1 reconciler.go:450] xxxxxx-sncmc-worker-us-east-1a-t72sb: Failed to find existing instance by id i-0ca2e1f7a004dbc7f: InvalidInstanceID.NotFound: The instance ID 'i-0ca2e1f7a004dbc7f' does not exist
        status code: 400, request id: cab0c58b-c3ef-4f7c-a295-a31f2ea656cf
I1005 23:05:28.471091       1 reconciler.go:261] xxxxxx-sncmc-worker-us-east-1a-t72sb: Possible eventual-consistency discrepancy; returning an error to requeue
E1005 23:05:28.471115       1 controller.go:246] xxxxxx-sncmc-worker-us-east-1a-t72sb: failed to check if machine exists: requeue in: 20s

Version-Release number of selected component (if applicable):
4.8.11

Expected results:
The machine should be deleted if the instance no longer exists.