Support for accelerated computing instance types in Lambda Managed Instances

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
Lambda Managed Instances currently provides access to CPU-based EC2 instances, enabling customers to leverage specialized hardware configurations while maintaining Lambda's serverless simplicity. (https://docs.aws.amazon.com/lambda/latest/dg/lambda-managed-instances.html)
We propose expanding the available instance types to include accelerated computing instances for workloads such as machine learning inference, media processing, graphics rendering, scientific computing, and data analytics. This capability would allow teams to optimize performance for parallel processing, matrix operations, and high-throughput computational functions while maintaining Lambda's serverless operational model—event source integration, automatic scaling, managed infrastructure, and simplified deployment.