Preparing an alternative high-level IR execution framework for the training code

[ememos]

It seems necessary to switch from directly executing the low-level nodes of the IR graph transformed with torch.fx to executing the high-level GraphModule processed through split_module(). The reason for this is to address the relatively slow training of GPT-like models in the low-level execution framework, even though the training of composite models is fast.

[ememos]

Added the following file:

• fx_dist_pp_training_type-C_gpt2_gpu.py