Private OpenShift clusters should by default provision private LoadBalancers

1. Proposed title of this feature request
Private OpenShift clusters should provision private LoadBalancers by default

2. What is the nature and description of the request?

• There are customer who have a private OpenShift cluster provisioned on AWS. By “private”, it means that the OpenShift cluster can only be accessed from an internal network and that this cluster is not visible and reachable from the internet. Currently when installing operators or helm charts on a private OpenShift cluster could result in LoadBalancer IPs being publicly exposed.

• When customers create a Service of type LoadBalancer, it'll be assigned a public IP (eg. **********.eu-west-2.elb.amazonaws.com). This seems very dangerous behaviour for a private cluster as the LoadBalancers could expose a private cluster to the Internet. It is only a matter of time before a customer experiences a breach because of this surprising and undocumented behaviour.

• By default, we would expect it to provision a private ELB, just like the ingress controller, so that the customers can safely expose apps to only internal consumers.

• Ideally, there should be a configuration option for what kind of ELB to assign, which for a private cluster should by default be private to avoid any nasty unwanted security incidents.

3. Why does the customer need this? (List the business requirements here)
Customers need this so that they can safely use operators and helm charts without having to worry about accidentally exposing our private cluster to the Internet. Also customers that are unaware of this undocumented behaviour will not be put at risk of a security breach.

4. How would the customer like to achieve this? (List the functional requirements here)
LoadBalancer IPs should be automatically annotated appropriately in a private cluster to provision private IPs (this is actually the behaviour in the latest version of AWS' load balancer controller)