Fix llama.covert_onnx to make it runnable in CI

[mszhanyi]

Description

1. make parity_check use local model to avoid using hf token
2. del the model didn't work because it tried to del the object define out of the function scope.
   So it caused out of memory in A10.
3. In fact, 16G GPU memory (one T4) is enough. But the conversion process always be killed in T4 and it works on A10/24G.
   Standard_NC4as_T4_v3 has 28G CPU memory
   Standard_NV36ads_A10_v5 has 440G memory.
   It looks that the model conversion needs very huge memory.

Motivation and Context

Last time, I came across some issues in convert_to_onnx.py so I use the onnx model in https://github.com/microsoft/Llama-2-Onnx for testing.
Now, these issues could be fixed. So I use onnx model generated by this repo and the CI can cover the model conversion.

[kunal-vaishnavi]

Now that torch v2.2.0 has been released in stable, can you also update the below lines to say torch>=2.2.0?

onnxruntime/onnxruntime/python/tools/transformers/models/llama/requirements.txt

Line 3 in eb0ce86

torch>=2.2.0.dev20230920

onnxruntime/onnxruntime/python/tools/transformers/models/llama/requirements-cuda.txt

Line 2 in eb0ce86

# Please manually install torch>=2.2.0.dev20230920 with CUDA enabled for the CUDA version installed in your system.