Add support for aggregating metrics collected by Project Prometheis to a central Thanos or Cortex configuration

[aiyengar2]

Is your feature request related to a problem? Please describe.

Currently, Prometheus Federator is designed to have a central Cluster Prometheus federate metrics down to all the underlying Project Prometheis, which can then collect custom metric workloads across all Projects in the cluster. However, a user may want to have all of those project-specific metrics then collected on a global level across one or more clusters.

Describe the solution you'd like

rancher-project-monitoring should support setting the remoteWrite and remoteRead configurations on the Project Prometheis with docs that explain how to configure it such that the metrics are shipped to a central Thanos / Cortex deployment.

As a result of the valuesOverride option, a Cluster Admin should also be able to configure all Project Prometheuses to point at a single central Thanos / cortex deployment by default, without allowing Project Owners to modify these settings.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.