Refactor evaluation module and add unit tests

[Copilot]

Refactored evaluation functionality into a dedicated auto_round/eval module and added comprehensive unit tests for CPU and GPU backends.

Changes

Module Structure

• auto_round/eval/eval_cli.py (486 lines): CLI argument parsing, vLLM/HF backend orchestration, custom argument parsing
• auto_round/eval/evaluation.py (439 lines): Core evaluation wrappers for lm_eval library
• auto_round/eval/__init__.py: Module exports

Unit Tests

• test/test_cpu/advanced/test_evaluation_functions.py (121 lines): Tests for parse_vllm_args() with various types (int, float, bool, string, mixed)
• test/test_cuda/advanced/test_evaluation.py (115 lines): Integration tests for vLLM backend with custom args, quantization workflows

Compatibility & Optimization

• Added allow_deprecated_quantization: True for vLLM 0.14.0 compatibility
• Removed redundant test cases to reduce CI execution time
• Added max_model_len=2048 parameter to vLLM tests for faster execution

Example Usage

from auto_round.eval.eval_cli import parse_vllm_args

# Parse custom vLLM arguments
args = parse_vllm_args("--tensor_parallel_size=2,--gpu_memory_utilization=0.9")
# Returns: {'tensor_parallel_size': 2, 'gpu_memory_utilization': 0.9}

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.