make gptq convert work for moe

[vkuzo]

Summary:

Makes gptq + moe + nvfp4 work e2e, results as expected on tiny model +
tiny dataset

Test Plan:

> TRITON_ALLOW_NON_CONSTEXPR_GLOBALS=1 torchao/prototype/gptq/gptq_nvfp4_olmoe_1b_7b_nonsequential_wikitext.sh

bf16
| Tasks  |Version|Filter|n-shot|    Metric     |   |Value |   |Stderr|
|--------|------:|------|-----:|---------------|---|-----:|---|------|
|wikitext|      2|none  |     0|bits_per_byte  |↓  |0.5895|±  |   N/A|
|        |       |none  |     0|byte_perplexity|↓  |1.5047|±  |   N/A|
|        |       |none  |     0|word_perplexity|↓  |8.8910|±  |   N/A|
real    0m59.219s
user    0m42.554s
sys     0m20.534s

nvfp4-rtn

| Tasks  |Version|Filter|n-shot|    Metric     |   |Value |   |Stderr|
|--------|------:|------|-----:|---------------|---|-----:|---|------|
|wikitext|      2|none  |     0|bits_per_byte  |↓  |0.6024|±  |   N/A|
|        |       |none  |     0|byte_perplexity|↓  |1.5183|±  |   N/A|
|        |       |none  |     0|word_perplexity|↓  |9.3277|±  |   N/A|
real    0m42.528s
user    0m41.217s
sys     0m12.817s

nvfp4-nonsequential with 4096 calibration samples on c4
| Tasks  |Version|Filter|n-shot|    Metric     |   |Value |   |Stderr|
|--------|------:|------|-----:|---------------|---|-----:|---|------|
|wikitext|      2|none  |     0|bits_per_byte  |↓  |0.6019|±  |   N/A|
|        |       |none  |     0|byte_perplexity|↓  |1.5177|±  |   N/A|
|        |       |none  |     0|word_perplexity|↓  |9.3087|±  |   N/A|
real    22m28.505s
user    22m36.008s
sys     0m13.872s

[vkuzo]

Stack from ghstack (oldest at bottom):

• -> make gptq convert work for moe #4330

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/4330

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Rolling out OSDC (ARC) runners on pull & trunk workflows in PyTorch main

⏳ 11 Pending, 1 Unrelated Failure

As of commit 65c616d with merge base 1c5e71e ():

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• Run 1xL4 Tests / test (SM-89, linux.g6.4xlarge.experimental.nvidia.gpu, --pre torch --index-url https://download.p... / linux-job (gh) (matched linux rule in flaky-rules.json)
  The process '/usr/bin/git' failed with exit code 128

This comment was automatically generated by Dr. CI and updates every 15 minutes.