Provide example on how to bootstrap a self-managed cluster

[anmazzotti]

What this PR does / why we need it:

This is a first attempt at providing a self-managed Rancher cluster.
The goal is to use a throwaway CAPI management cluster to provision a downstream Rancher cluster and pivot to it.

Unfortunately the move operation is currently not working.
clusterctl does not seem to recognize the providers installed by the cluster-api-operator and fails with:

> clusterctl move --to-kubeconfig=/tmp/cluster-api/rancher
Performing move...
Error: failed to check providers in target cluster: [provider infrastructure-docker not found in the target cluster, provider cluster-api not found in the target cluster, provider bootstrap-rke2 not found in the target cluster, provider control-plane-rke2 not found in the target cluster]

Note that the move operation succeeds when instead the downstream cluster is also initialized via clusterctl init, however this scenario is not suitable for us, as the requirement is to provision a Rancher Turtles ready environment instead.

The solution is to implement the clusteroperator move command.

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #984

Special notes for your reviewer:

Checklist:

• squashed commits into logical changes
• includes documentation
• adds unit tests
• adds or updates e2e tests

[anmazzotti]

Closing this pull request since the proof of concept has been translated to an ADR