Provide a pattern to wait for RBAC propagation

We have multiple services that hit a problem where their live tests are failing because RBAC permissions were not propagated in time. We also had many ways each of them tries to solve the problem:

By waiting pre-defined amount of time: https://github.com/Azure/azure-sdk-for-net/blob/master/sdk/eventhub/test-resources-post.ps1?rgh-link-date=2021-04-21T00%3A02%3A41Z#L15-L16

Or having retry on tests: https://github.com/Azure/azure-sdk-for-net/pull/20559

To me, the goal of `New-TestResources` is to leave the test environment fully ready for the test run and non-propagated RBAC is something we should solve in a centralized way.

Start by reading the New-TestResources flow and the linked test-resources-post.ps1 lines, then compare the retry approach discussed in pull request 20559. Define how centralized RBAC propagation handling should leave the test environment ready before live tests run; done means the affected services no longer need separate waits or test retries.

Written by the indexing model from the issue text.