format add check for scheme args

[n1ck-guo]

Description

In the previous code, the format function only handles cases where the scheme is a string. It does not perform checks for scenarios like --scheme w2a16 --bits 4. This PR adds the relevant checks.

Type of Change

• Bug fix
• New feature
• Documentation update
• Performance improvement
• Code refactoring
• Other (please specify):

Related Issues

Fixes or relates to #1247

Checklist Before Submitting

• My code has been tested locally.
• Documentation has been updated as needed.
• New or updated tests are included where applicable.

[Copilot]

Pull request overview

This PR adds validation of scheme arguments when exporting models to different quantization formats. Previously, the format function only validated string schemes and didn't check for incompatibilities between the scheme object's properties (bits, data_type, etc.) and the target format's requirements.

Changes:

• Added check_scheme_args() method to validate QuantizationScheme objects against format-specific constraints
• Implemented validation for LLMCompressor, AutoGPTQ, AutoAWQ, and VLLM formats
• Removed unsupported w8a8 test case for llmcompressor format

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
auto_round/formats.py	Added check_scheme_args() methods to OutputFormat classes to validate scheme parameters against format requirements
test/test_cpu/export/test_export.py	Added test cases to verify scheme validation raises appropriate errors for incompatible configurations
test/test_cpu/integrations/test_llmcompressor.py	Commented out w8a8 test case as this scheme is not supported by llmcompressor format

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.