Support for AsyncLLMEngine

[agunapal]

Hello,

I believe AsyncLLMEngine was removed sometime ago.

I was wondering if there is support for loading a built engine and support asynchronous inference requests.

I have seen that we have this https://github.com/NVIDIA/TensorRT-LLM/blob/main/tests/hlapi/test_llm.py#L288 but this is not for the built engine?

[agunapal]

I have verified that model can be LLMEngine.