allow backup / restore to restore to a cluster with different topology

The backup and restore utility requires that you restore to an environment with the same topology (i.e. number pms and dbroots) as the original environment. In some cases it would be desirable to have an ability to restore to an environment that has a different topology either because this is what you have or you want to clone the production instance to a smaller test environment. I suspect downsizing is probably the more common use case but i could also see the inverse being valid where you want to migrate to a larger cluster.

Challenges with this would be you would need to support a mapping to instruct where dbroots and pm end up at and finally to update the extent map to reflect these changes. I don't think we'd want to allow for any online re-organization or merging of dbroots as part of this as theoretically this could be handled after with the data redistribution logic.