Add tests/gpu_vllm (#1517)

### What does this PR do?

Type of change: new tests

This PR adds unit tests for vLLM fakequant, specifically testing code in
`modelopt/torch/quantization/plugins/vllm.py`


### Testing

```
pytest tests/gpu_vllm/torch/quantization/test_vllm_dynamic_modules.py -sv
```

### Before your PR is "*Ready for review*"

Make sure you read and follow [Contributor
guidelines](https://github.com/NVIDIA/Model-Optimizer/blob/main/CONTRIBUTING.md)
and your commits are signed (`git commit -s -S`).

Make sure you read and follow the [Security Best
Practices](https://github.com/NVIDIA/Model-Optimizer/blob/main/SECURITY.md#security-coding-practices-for-contributors)
(e.g. avoiding hardcoded `trust_remote_code=True`, `torch.load(...,
weights_only=False)`, `pickle`, etc.).

- Is this change backward compatible?: ✅ 
- If you copied code from any other sources or added a new PIP
dependency, did you follow guidance in `CONTRIBUTING.md`: N/A
- Did you write any new necessary tests?: ✅ 
- Did you update
[Changelog](https://github.com/NVIDIA/Model-Optimizer/blob/main/CHANGELOG.rst)?:
N/A
- Did you get Claude approval on this PR?: ✅ 

### Additional Information


<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **Tests**
* Added comprehensive GPU vLLM test suite with end-to-end quantization
checks and fixtures for TinyLlama, TinyQwen3-MoE, and DeepSeek V3;
includes helpers to build tiny DeepSeek V3 models.

* **Chores**
* Updated GPU CI to use explicit container image references, added a
GPU-focused test session, and adjusted test-run setup for vLLM.

* **Documentation**
  * Documented new GPU test directory in contributing guide.

<!-- review_stack_entry_start -->

[![Review Change
Stack](https://storage.googleapis.com/coderabbit_public_assets/review-stack-in-coderabbit-ui.svg)](https://app.coderabbit.ai/change-stack/NVIDIA/Model-Optimizer/pull/1517?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack)

<!-- review_stack_entry_end -->
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Signed-off-by: Kinjal Patel <kinjalpravin@nvidia.com>
Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>
Co-authored-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>

.github/workflows/_example_tests_runner.yml

@@ -34,9 +34,6 @@ jobs:
     timeout-minutes: ${{ inputs.timeout_minutes }}
     container:
       image: ${{ inputs.docker_image }}
-      credentials:
-        username: $oauthtoken
-        password: ${{ secrets.NGC_API_KEY }}
       options: --shm-size=2gb # TRT-LLM tests on 2-GPU runner needs more shared memory
       env:
         PIP_CONSTRAINT: "" # Disable pip constraint for upgrading packages

.github/workflows/gpu_tests.yml

@@ -29,6 +29,7 @@ jobs:
         tests/gpu/**
         tests/gpu_megatron/**
         tests/gpu_trtllm/**
+        tests/gpu_vllm/**
 
   gpu-tests:
     needs: [pr-gate]
@@ -39,25 +40,30 @@ jobs:
         include:
           - example: gpu
             timeout: 75
-            container_image: pytorch:26.04-py3
+            container_image: nvcr.io/nvidia/pytorch:26.04-py3
           - example: gpu_megatron
             timeout: 45
-            container_image: nemo:26.04
+            container_image: nvcr.io/nvidia/nemo:26.04
           - example: gpu_trtllm
             timeout: 30
-            container_image: tensorrt-llm/release:1.3.0rc16
+            container_image: nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc16
+          - example: gpu_vllm
+            timeout: 30
+            container_image: docker.io/vllm/vllm-openai:v0.20.0
     runs-on: ${{ startsWith(github.ref, 'refs/heads/pull-request/') && 'linux-amd64-gpu-rtxpro6000-latest-1' || 'linux-amd64-gpu-rtxpro6000-latest-2' }}
     timeout-minutes: ${{ matrix.timeout }}
     container:
-      image: nvcr.io/nvidia/${{ matrix.container_image }}
-      credentials:
-        username: $oauthtoken
-        password: ${{ secrets.NGC_API_KEY }}
+      image: ${{ matrix.container_image }}
       env:
         GIT_DEPTH: 1000 # For correct version
         PIP_CONSTRAINT: "" # Disable pip constraint for upgrading packages
         HF_TOKEN: ${{ secrets.HF_TOKEN }}
     steps:
+      - name: Install git
+        # The vllm container ships without git; needed for a real checkout (correct
+        # setuptools-scm version) and for the Codecov upload below.
+        if: matrix.example == 'gpu_vllm'
+        run: apt-get update && apt-get install -y git
       - uses: actions/checkout@v6
       - uses: nv-gha-runners/setup-proxy-cache@main
       - name: Setup environment variables
@@ -68,7 +74,8 @@ jobs:
           COVERAGE_PROCESS_START: ${{ github.workspace }}/pyproject.toml
           COVERAGE_FILE: ${{ github.workspace }}/.coverage
         run: |
-          python -m pip install nox && nox -s ${{ matrix.example }}
+          # Use `python3` (the vllm image has no `python` on PATH)
+          python3 -m pip install nox && nox -s ${{ matrix.example }}
       - name: Upload GPU coverage to Codecov
         uses: codecov/codecov-action@v5
         with: