fix: simplify configuration for baremetal hub and spoke

[butler54]

Consider the following scenarios:

1. I have a hub cluster for baremetal. It could have no tee support, intel or amd.
2. Consider I have spoke managed clusters. They target the spoke cluster group. Each cluster could have intel or AMD (mixed TEE's in one workload cluster is not supported).

The objective here is to minimize the number of configuration points.