Refactor evaluation in tests to use evaluate_accuracy function

[xin3he]

Description

Refactor evaluation in tests to use evaluate_accuracy function

• Replaced instances of simple_evaluate_user_model with evaluate_accuracy across multiple test files.
• Updated accuracy thresholds for various tasks in test cases to ensure consistency.
• Removed redundant imports and functions related to accuracy evaluation.
• Enhanced readability and maintainability of the test code by centralizing accuracy evaluation logic.

Type of Change

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

Related Issues

Fixes or relates to #

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 refactors test files to use a centralized evaluate_accuracy helper function, replacing multiple instances of simple_evaluate_user_model and simple_evaluate calls. The changes improve code maintainability by consolidating evaluation logic and removing redundant helper functions.

Changes:

• Replaced direct evaluation calls with the evaluate_accuracy helper function across test files
• Removed duplicate helper functions (get_accuracy) from multiple test files
• Updated import statements to include evaluate_accuracy from the helpers module
• Cleaned up unused imports including re, simple_evaluate, and simple_evaluate_user_model

Reviewed changes

Copilot reviewed 22 out of 22 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
test/helpers.py	Added centralized evaluate_accuracy and generate_prompt helper functions
test/test_cuda/utils/test_alg_ext.py	Replaced evaluation call with evaluate_accuracy helper
test/test_cuda/schemes/test_auto_scheme.py	Updated to use evaluate_accuracy helper
test/test_cuda/quantization/test_mxfp_nvfp.py	Replaced evaluation with evaluate_accuracy helper
test/test_cuda/quantization/test_mix_bits.py	Updated to use evaluate_accuracy helper
test/test_cuda/quantization/test_asym.py	Removed unused imports
test/test_cuda/quantization/test_2_3bits.py	Replaced evaluation calls and removed duplicate get_accuracy function
test/test_cuda/export/test_gguf.py	Updated to use evaluate_accuracy helper
test/test_cuda/export/test_auto_round_format.py	Replaced evaluation calls with evaluate_accuracy helper
test/test_cuda/core/test_main_func.py	Updated evaluation calls and removed duplicate get_accuracy function
test/test_cuda/backends/test_triton_backend.py	Replaced evaluation calls with evaluate_accuracy helper
test/test_cuda/backends/test_torch_backend.py	Updated to use evaluate_accuracy helper
test/test_cuda/backends/test_marlin_backend.py	Replaced evaluation calls with evaluate_accuracy helper
test/test_cuda/backends/test_exllamav2_backend.py	Updated to use evaluate_accuracy helper
test/test_cuda/advanced/test_multiple_card_calib.py	Removed duplicate get_accuracy function
test/test_cuda/advanced/test_multiple_card.py	Removed duplicate get_accuracy function and updated evaluation calls
test/test_cuda/advanced/test_fp8_input.py	Moved helper functions to centralized location
test/test_cpu/quantization/test_mix_bits.py	Updated to use evaluate_accuracy helper
test/test_cpu/quantization/test_asym.py	Removed unused imports
test/test_cpu/core/test_autoround.py	Replaced evaluation calls with evaluate_accuracy helper
test/test_cpu/backends/test_torch_backend.py	Updated to use evaluate_accuracy helper
test/test_ark/test_model.py	Replaced evaluation call with evaluate_accuracy helper

[xin3he]

Let‘s target it to 1.0.0