Feature-90 minute timeout for on-demand capacity mode

We would like to be able to run functions for longer than 15 minutes in on-demand mode. 90 minutes would be parity with the Lambda managed instances mode recently released.

**Note**
> * Please vote on this issue by adding a 👍 reaction 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 providing additional feedback, please leave a comment

**Description**

see blog post:

https://aws.amazon.com/blogs/compute/announcing-90-minute-function-timeout-on-aws-lambda-managed-instances/

**Additional context**

We still have several jobs that must run on ECS because it is either too costly to rearchitect them for 15 minute time outs or impractical due to the unit of work requiring more time. We would value a single place to run our jobs rather than having some on Lambda and some on ECS. We are much less interested in adding LMI as yet another variant in our platform.

Start by reviewing the requested 90-minute timeout and the linked AWS Lambda managed-instances blog post; the payload identifies no repository files, tests, or implementation entry points. Done means defining and delivering on-demand capacity support for function runs up to 90 minutes, with behavior matching the requested parity.

Written by the indexing model from the issue text.