Using the Cloud Pak Deployer on a cluster that already has the NFD or OpenShift AI operator will fail if their install value is true

[David-M-IBM]

Describe the bug
It looks like if the NFD operator or OpenShift AI operator are already installed it will prevent the cloud pak deployer from progressing if it is set to openshift: gpu: install: true or openshift: openshift_ai: install: true

This occurs due to the fact that it creates another operator group in the existing namespace and that causes the CSVs status to become Failed because there can't be more than one operator group. I'm not sure if this is an issue that matters to anyone else, but in situations where they're already installed you'd need to provide a value of false for the install to progress.

I'm not sure if this is even a problem as you can just change the values to false, but with the fact that more and more cp4d components require these operators, users might encounter this when trying to use the deployer.

To Reproduce
Steps to reproduce the behavior:

1. Install the NFD or OpenShift AI operator before kicking off the cloud pak deployer.
2. Try to run the cloud pak deployer job with openshift: gpu: install: true or openshift: openshift_ai: install: true
3. It will loop trying to verify the CSV has a status of Successful but it will be failed because of a conflict of two operator groups existing in the same namespace
   cloud-pak-deployer-csmwk-cloud-pak-deployer.log

Expected behavior
It might make sense to do the CSV check before applying the yaml for the operators and if it already exists and is already successful then just skip the step. It might be more annoying trying to verify that it is the right version though.