Spawned from: https://coreos.slack.com/archives/CFN0XJEPR/p1644929623040169

We need updated documentation that has our recommendation for how users should handle APIs that are no longer present on their destination clusters, i.e. a 3.11 to 4.10 migration. We have an old blog posted here that we can use as a starting point: https://cloud.redhat.com/blog/migrating-openshift-apps

Questions we have to answer:

• How does MTC handle version mismatch?
• Once a version mismatch is detected, how do I handle that so I can still migrate my workload?

Our current documentation around this is lacking, and its growing in importance as OCP continues to move further from OCP 3.11 and k8s 1.11.

DOD: Upstream document merged that answers the above questions.