Add support for a Python vLLM backend

[rgreenberg1]

Description:
Right now, GuideLLM only works w/ GuideLLM through the OpenAI compatible API backend to take the target URI and perform benchmarking on a vLLM server. We want to also add support for the python vLLM backend so that we can create scripts for benchmarks w/ vLLM and use it in-line with Python.

Acceptance Criteria:

• Add support for the Python vLLM Backend
• Potentially helpful: https://github.com/triton-inference-server/vllm_backend - maybe not