Description of problem:

By default the cluster gets set up on AWS with the following DNS structure:

```
Wildcard Apps DNS:  *.apps.<clustername>.<basedomain>
API DNS: api.<clustername>.<basedomain>
ETCD: etcd-0.<clustername>.<basedomain>
```

As we use separate basedomains for our environments (exampledev.com / example.com) we would rather want to create the cluster without an additional clustername, meaning we want to achieve this DNS structure:

```
Wildcard Apps DNS: *.<basedomain>
API DNS: api.<basedomain>
ETCD: etcd-0.<basedomain>
```

So we get our console to be reachable by default under

```
https://console-openshift-console.exampledev.com 
```
and not

```
https://console-openshift-console.apps.myclustername.exampledev.com
```

But we can't do this as the clustername is not optional and must be included into the install-config.yaml.


We tried changing the basedomain after the installation by modifying the `ingress.config.openshift.io/cluster` but then the ingresscontroller tried to create a *.<basedomain> entry in the DNS Private HostedZone <clustername>.<basedomain> on AWS and got this error:

```
2019-06-26T08:54:33.742Z        ERROR   operator.init.controller-runtime.controller     controller/controller.go:217    Reconciler error        {"controller": "operator-controller", "request": "openshift-ingress-operator/default", "error": "failed to ensure ingresscontroller: failed to ensure DNS for default: failed to ensure DNS record &{{ map[Name:test-azaa-int kubernetes.io/cluster/test-cm5ql:owned]} ALIAS *.exampledev.com -> ###.eu-central-1.elb.amazonaws.com} for openshift-ingress-operator/default: failed to update alias in zone Z065998772O1YGQ3K653: couldn't update DNS record in zone Z06###653: InvalidChangeBatch: [RRSet with DNS name \\052.exampledev.com. is not permitted in zone test.exampledev.com.]\n\tstatus code: 400, request id: 04963a50-97f0-11e9-8cee-1b9d0c13ce08", "errorCauses": [{"error": "failed to ensure ingresscontroller: failed to ensure DNS for default: failed to ensure DNS record &{{ map[Name:test-cm5ql-int kubernetes.io/cluster/test-cm5ql:owned]} ALIAS *.exampledev.com -> ###.eu-central-1.elb.amazonaws.com} for openshift-ingress-operator/default: failed to update alias in zone Z06###653: couldn't update DNS record in zone Z06###653: InvalidChangeBatch: [RRSet with DNS name \\052.exampledev.com. is not permitted in zone test.exampledev.com.]\n\tstatus code: 400, request id: 049###e08"}]}
```

Is there a way to create a cluster directly reachable on the basedomain without any clustername prefix? If not can this be added as an option? 



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

```
$ openshift-install version
bin/openshift-install v4.1.0-201905212232-dirty
built from commit 71d8978039726046929729ad15302973e3da18ce
release image quay.io/openshift-release-dev/ocp-release@sha256:b8307ac0f3ec4ac86c3f3b52846425205022da52c16f56ec31cbe428501001d6
```

How reproducible:

Install a cluster on AWS. Try to change the basedomain of the cluster from `apps.<clustername>.<basedomain>` to `<basedomain>`.


Steps to Reproduce:
1. Install OCP 4.1 Cluster on AWS with openshift-installer
2. After installation `oc edit ingress.config.opneshift.io/cluster`
3. Change spec.domain and remove apps.<clustername>. from it.

Actual results:

Ingress-Operator writes error as it can't update the DNS on the private Zone <clustername>.<basedomain>, as it can't add a *.<basedomain> entry into it.


Expected results:


Cluster can be set up directly on basedomain rather than using a separate clustername prefix as subdomain.


Additional info:

We already use this DNS structure on our OCP 3.11 installations on AWS and want to sidegrade to 4.1 and keep this DNS structure.