Fail early in the API if the release image is not multi-arch and the management cluster's CPU architecture is not the same as the NodePool's CPU architecture

User Story:

As a user of the HyperShift, I would like the API to fail early if these conditions are all true:

• the release image is not multi-arch
• the management cluster's CPU architecture is not the same as the NodePool's CPU architecture

so that we can prevent a HostedCluster from continuing to be created that will have errors due mismatches between the release image, management cluster's CPU architecture, and NodePool's CPU architecture.

Acceptance Criteria:

• The HyperShift API fails to create a cluster when the release image is not multi-arch and the management cluster's CPU architecture does not match the NodePool's CPU architecture.
• There is documentation providing information the API will fail when it meets the conditions above.

Out of Scope:

This should be done for the CLI as well but will be covered thru HOSTEDCP-1104.

Engineering Details:

• HyperShift CLI currently defaults to a multi-arch image in version.go.

This requires/does not require a design proposal.
This requires/does not require a feature gate.