[ECS] [bug]: setting startTimeout above 3 hours has unexpected results

### 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

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

**Steps to reproduce**
Define a task T with 2 containers, A and B. A is a container which sleeps for 4 hours:
`{"containerName": "A", "essential": false, "command": ["sleep", "14400"], ...`

B is a container with a dependency on A and a `startTimeout` of more than 4 hours:
`{"dependsOn": {"containerName" :"A", "condition": "COMPLETE"}, "startTimeout": 15000,...`

Schedule an instance of task T.

**Expected behavior**

In order of preferability, either:
* Container B should be `PENDING` for 4 hours until A completes, at which point B runs.
* Setting startTimeout greater than 3 hours (which seems to be controlled by `ECS_ENGINE_TASK_CLEANUP_WAIT_DURATION`?) should produce an error when defining the task.

**Observed Behavior**

The task is stopped after 3 hours with the following message:

> task last status remained in PENDING too long

**Additional context**

Reproduced on Fargate 1.4.0

Start by reviewing the ECS task definition and Fargate 1.4.0 reproduction: container A sleeps for 4 hours, while dependent container B has a startTimeout of 15000. Compare the observed three-hour stop with the two proposed expected behaviors; done would require an agreed ECS behavior or validation outcome, but no repository files or tests are identified.

Written by the indexing model from the issue text.