Add LM Eval Testing

[dsikka]

Summary:

• Add LM Eval Teseting to run on a weekly cadence
• Model is quantized after which lm_eval runs and the metrics produced are asserted against metrics defined in the config for the particular test case
• Updates the bash script to take in configs folder and test script, allowing it to work for both the vLLM e2e tests and LM Eval tests

Testing

• Testing has to be updated to use the following commands:

bash tests/e2e/vLLM/run_tests.sh -c tests/e2e/vLLM/lm_eval_configs -t tests/e2e/vLLM/test_lmeval.py 
bash tests/e2e/vLLM/run_tests.sh -c tests/e2e/vLLM/configs -t tests/e2e/vLLM/test_vllm.py 

• LM Eval testing requires CADENCE=weekly

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

[dsikka]

• To add in future PRs:
• w4 + act order
• int8 + smoothquant

[horheynm]

one ocmment otherwise lgtm

[rahul-tuli]

LGTM pending Domenic's comment about "usage" function in the bash script