Sometimes deleting the bootstrap ssh rule during bootstrap destroy can timeout after 5min, failing the installation.

    4.16+ with capi/aws

    Intermittent

    1.
    2.
    3.
    

                level=info msg=Waiting up to 5m0s (until 2:31AM UTC) for bootstrap SSH rule to be destroyed...
                level=fatal msg=error destroying bootstrap resources failed during the destroy bootstrap hook: failed to remove bootstrap SSH rule: bootstrap ssh rule was not removed within 5m0s: timed out waiting for the condition

    The rule is deleted successfully and in a timely manner.

    This is probably happening because we are changing the AWSCluster object, thus causing capi/capa to trigger a big reconciliation of the resources. We should try to delete the rule via aws sdk.