[Refactor][Config] Remove weight NZ environment variable

[kk-ss1999]

What this PR does / why we need it?

Removes the legacy VLLM_ASCEND_ENABLE_NZ environment variable after its behavior was migrated to additional_config.weight_nz_mode. Runtime code, deployment examples, test configurations, and documentation are migrated to the config field where applicable.

This is one independently reviewable part of the seven-variable split of #14637.

Does this PR introduce any user-facing change?

Yes. VLLM_ASCEND_ENABLE_NZ is no longer read. Users must configure the behavior with --additional-config '{"weight_nz_mode":1}'.

How was this patch tested?

• ruff check and ruff format --check on changed Python files

• Python byte-compilation of changed Python files

• YAML parsing of changed YAML files

• JSON parsing of concrete additional-config payloads

• Scan confirming no active references to VLLM_ASCEND_ENABLE_NZ remain outside migration documentation and the ignored-variable unit test

• NPU end-to-end tests were not run in this Windows environment

• vLLM version: v0.27.1

• vLLM main: vllm-project/vllm@58d3918

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request completes the migration of the VLLM_ASCEND_ENABLE_NZ environment variable to the additional_config.weight_nz_mode configuration field. This change standardizes configuration management for Ascend-based deployments, ensuring that users move away from legacy environment variables in favor of structured configuration payloads. The update impacts runtime logic, test suites, and documentation, providing a cleaner and more consistent interface for configuring weight NZ modes.

Highlights

• Environment Variable Removal: Removed the legacy VLLM_ASCEND_ENABLE_NZ environment variable, completing its migration to the additional_config.weight_nz_mode configuration field.
• Configuration Migration: Updated runtime code, deployment documentation, and test configurations to use the additional_config parameter for setting weight NZ mode.
• Documentation Updates: Updated user guides to clarify that VLLM_ASCEND_ENABLE_NZ is no longer supported and provided instructions for the new configuration approach.

New Features

🧠 You can now enable Memory (public preview) to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize the Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

Suggested PR Title:

[Ops][Misc] Remove legacy VLLM_ASCEND_ENABLE_NZ environment variable and migrate to weight_nz_mode

Suggested PR Summary:

### What this PR does / why we need it?
This PR removes the legacy `VLLM_ASCEND_ENABLE_NZ` environment variable and fully transitions to using `weight_nz_mode` within `additional_config` for configuring the weight NZ mode on Ascend NPU. It updates documentation, examples, end-to-end tests, and unit tests to reflect this change.

Feedback: In `tests/e2e/weekly/single_node/models/test_qwen3_30b_acc.py`, the removal of `VLLM_ASCEND_ENABLE_NZ="2"` was not migrated to `weight_nz_mode: 2` in the server arguments, which may cause accuracy degradation or test failures.

### Does this PR introduce _any_ user-facing change?
Yes, the legacy `VLLM_ASCEND_ENABLE_NZ` environment variable is no longer supported. Users must now use `additional_config.weight_nz_mode` (with values `0` to disable NZ, `1` to enable NZ only for quantized weights, and `2` to enable NZ for BF16/FP16 weights when supported).

### How was this patch tested?
The changes were tested by updating existing unit tests and various end-to-end test configurations.

[gemini-code-assist]

The environment variable VLLM_ASCEND_ENABLE_NZ was set to "2" in this test, but it has been removed without migrating the configuration to --additional-config '{"weight_nz_mode": 2}' in server_args. This will cause the test to run with the default weight_nz_mode of 1, which might lead to accuracy degradation or test failures for the Qwen3-30B-A3B-W8A8 model. Please ensure "weight_nz_mode": 2 is added to the --additional-config argument in server_args.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.