[ECS] [request]: Evacuate and Fence an Availability Zone at runtime

### Community Note

* Please vote on this issue by adding a 👍 [reaction](https://blog.github.com/2016-03-10-add-reactions-to-pull-requests-issues-and-comments/) to the original issue to help the community and maintainers prioritize this request
* Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
* If you are interested in working on this issue or have submitted a pull request, please leave a comment

**Tell us about your request**
At runtime:
a) Provide the capability for a customer to safely Evacuate running tasks from an availability zone (evacuating an AZ)
b) Provide the capability for a customer to mark an availability zone as undesirable thereby preventing any new tasks being launched into the AZ (fencing an AZ)

**Which service(s) is this request for?**
ECS Fargate

**Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?**
Currently, if a service that runs tasks in multiple AZs is experiencing a grey failure in one of the AZs, the only way to evacuate the AZ is by redeploying the service and its tasks with the AZ disabled/removed. This can be challenging to manage across hundreds of accounts.

**Are you currently working around this issue?**
No

No repository files, tests, or implementation entry points are identified in the request. Start by clarifying the ECS Fargate API and safety criteria for evacuating and fencing an Availability Zone; done means customers can perform both runtime operations without a service redeployment.

Written by the indexing model from the issue text.