Complete Google-style docstrings across neural_compressor

[Copilot]

This pull request primarily improves documentation and code clarity across the codebase, especially for the JAX quantization APIs and the core configuration and utility classes. It introduces new API documentation for JAX quantization, adds or enhances docstrings for many methods, and clarifies function signatures and argument types. Additionally, it fixes a typo in a function name and improves the descriptions of configuration and operator handling.

Documentation and API enhancements:

• Added JAX quantization API documentation to the API docs, including new files jax_quantization_common.rst and jax_quantization_config.rst and corresponding references in the main API index. [1] [2] [3]
• Added module-level docstrings for JAX quantization algorithm entry points and public API exports (neural_compressor/jax/algorithms/dynamic.py, neural_compressor/jax/algorithms/static.py, neural_compressor/jax/quantization/__init__.py). [1] [2] [3]
• Expanded or added detailed docstrings to many core methods and classes across configuration, tuning, and utility modules for better clarity and maintainability (e.g., BaseConfig, Evaluator, utility decorators, and logging functions). [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

JAX quantization API and config improvements:

• Improved function signatures and docstrings for dynamic_quantize and static_quantize to clarify argument types and return values. [1] [2]
• Enhanced the OperatorConfig class and related config registration methods with more descriptive docstrings and clarified return types in neural_compressor/jax/quantization/config.py. [1] [2] [3] [4]

Bugfixes and minor code improvements:

• Fixed a typo in the Evaluator class (molde → model).

These changes collectively improve the usability, maintainability, and clarity of the codebase and documentation, especially for users and developers working with JAX quantization features.

[chensuyue]

@copilot you have missing a list of issues such as

neural-compressor/neural_compressor/torch/algorithms/autoround/autoround.py

Line 243 in 786a272

split (str, optional): The data split to use. Defaults to None.

is not correct. Need to scan carefully in the path and fix the missing issues.

[Copilot]

@copilot you have missing a list of issues such as https://github.com/intel/neural-compressor/blob/786a2720e1f57f3f8ab09fcaf2ff585bc716a4ab/neural_compressor/torch/algorithms/autoround/a...

Updated autoround.py docstrings to align with actual parameters/defaults in get_dataloader and get_mllm_dataloader (removed non-existent split, corrected batch_size defaults, and documented missing args). Commit: ae3394a.

[chensuyue]

@copilot revert the PR description, need an overall describe, not only just for the last change.

[Copilot]

@copilot revert the PR description, need an overall describe, not only just for the last change.

Updated the PR description checklist to reflect the overall JAX docstring work. Commit: 8e8a5e6.