[WIP] Use vllm transformers backend for pooling model runner.#752

maxdebayser wants to merge 8 commits intomaintorch-spyre/sendnn-inference:mainfrom

vllm_model_loader_poolingtorch-spyre/sendnn-inference:vllm_model_loader_poolingCopy head branch name to clipboard