[ECS]: a summary of placement

Odd request, but ECS placement and scaling is getting more and more complicated.

I've got a very poor sense for how all the placment/scale features fit together, and when I should be using which ones.

* capacity provider
* capacity provider strategy
* cluster
* definition compatibility
* launch type
* placement constraint
* placement strategy
* schedule strategy
* service
* service autoscaling
* task group

This is to say nothing of the underlying EC2 and Fargate platforms, with launch configurations, launch templates, autoscaling groups, placement groups, availability zones, dedicated instances, reserved instances, spot instances, spot fleets, etc.

A cohesive brief definition of each concept and its role would really help. Like a one-page reference guide.

Start by inventorying the ECS placement and scaling concepts listed in the issue, including capacity providers, clusters, launch types, placement strategies, services, and autoscaling. The work is done when those concepts and their roles are explained cohesively in a one-page reference guide; no files or tests are named.

Written by the indexing model from the issue text.