[MoE Refactor] Migrate MoeWNA16Method quantization method over to using the new MK oracle scheme.

[bnellnm]

Purpose

This PR is an fixed + updated version of #42647

Migrate MoeWNA16Method quantization method over to using the new MK oracle scheme.

Additional fixes

• Fix Triton backend weight conversion when original weights are from auto_gptq.
• Fix marlin zero point conversion code when original ZPs come from auto_gptq.

Also fixes broken quantization import when humming module is not installed. This was broken prior to #42647

cc @bedeks @yzong-rh

Test Plan

• pytest tests/compile/h100/test_startup.py::test_model_startup[kimi_k2.5]
• pytest tests/evals/gsm8k/test_gsm8k_correctness.py::test_gsm8k_correctness[Qwen1.5-MoE-W4A16-CT] --config-list-file=configs/models-small.txt
• lm-eval run --model vllm --tasks gsm8k --model_args="pretrained=Qwen/Qwen1.5-MoE-A2.7B-Chat-GPTQ-Int4" --num_fewshot 5 --batch_size 2048
• lm-eval run --model vllm --tasks gsm8k --model_args="pretrained=TheBloke/Mixtral_7Bx2_MoE-AWQ" --num_fewshot 5 --batch_size 2048

Test Result

pytest tests pass.

lm-eval results:

Qwen baseline

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.4640|±  |0.0137|
|     |       |strict-match    |     5|exact_match|↑  |0.3283|±  |0.0129|

Qwen PR

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|                                                                                   
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|                                                                                   
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.4549|±  |0.0137|                                                                                   
|     |       |strict-match    |     5|exact_match|↑  |0.3237|±  |0.0129|                                                                                   

Mixtral baseline

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.6929|±  |0.0127|
|     |       |strict-match    |     5|exact_match|↑  |0.6899|±  |0.0127|

Mixtral PR

|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.6952|±  |0.0127|
|     |       |strict-match    |     5|exact_match|↑  |0.6922|±  |0.0127|

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.

[mergify]

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @bnellnm.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

[mergify]

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @bnellnm.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork