Improve teardown reliability handling

[JimMadge]

:white_check_mark: Checklist

• [x] I have searched open and closed issues for duplicates.
• [x] This is a request for a new feature in the Data Safe Haven or an upgrade to an existing feature.
• [x] The feature is still missing in the latest version.
• [x] I have read through the documentation.
• [x] This isn't an open-ended question (open a discussion if it is).

:strawberry: Suggested change

Teardown may not be able to remove all resources. In that instance the CLI will give a non-zero return code.

We are not able to ensure that all resources are removed, that is out of our control. We might want to look at how often this happens, and if we can improve the handling/messaging.

:steam_locomotive: How could this be done?

• How many repeats of pulumi destroy do we attempt?
• Are there particular resources which are problematic
• Can we catch this and give the user a message like "Not all resources were removed, this isn't your fault, please run the command again"

[jemrobinson]

We currently retry on particular previously-encountered failures (where Azure doesn't correctly indicate that resource A can't be deleted until resource B has been). A better failure error message would certainly be a good idea.