Linting/Formatting of tests folder

[Snuffy2]

This pull request focuses on improving the quality, clarity, and maintainability of the test suite for the BaseToolParser class in the mlx-openai-server project. The changes include code formatting, refactoring of test logic, and enhanced documentation. Below is a breakdown of the key modifications:

Files Changed

• __init__.py (new file): Added a package-level docstring to describe the purpose of the tests package, including a brief overview and import statement for future extensibility.
• test_base_tool_parser.py: Major refactoring of the test file to improve readability, correctness, and coverage.

Key Improvements

Test Refactoring and Clarity:

• Replaced complex, artificially chunked test inputs (e.g., split by # or @@ delimiters) with more realistic, streaming-style chunks that simulate actual tool call parsing scenarios.
• Updated expected outputs to match the new input structure, ensuring tests accurately reflect parser behavior for complete tool calls and surrounding text.
• Enhanced test logic to handle remaining content after processing all chunks, providing better coverage of edge cases.

Assertion and Error Handling Enhancements:

• Switched from self.assertEqual to plain assert statements for simpler, more direct assertions.
• Introduced strict=True in zip operations to enforce exact length matching between outputs and expectations, improving error detection and reliability.
• Added type annotations (e.g., -> None) to method signatures for better static analysis and adherence to Python best practices.

Code Quality and Style:

• Added module-level docstrings and improved inline documentation.
• Updated imports with type ignore comments to handle potential import issues.
• Removed unnecessary boilerplate code (e.g., if __name__ == "__main__": unittest.main()).

Documentation:

• Introduced a package-level docstring in __init__.py to clarify the role of the test suite.

Impact

• Maintainability: Tests are now easier to understand and modify, with realistic inputs that better mimic real-world usage.
• Reliability: Improved assertions reduce false positives and provide clearer error messages during test failures.
• Compliance: Changes align with the project's guidelines for code quality, type annotations, and documentation (as outlined in AGENTS.md).
• Coverage: Enhanced handling of streaming scenarios ensures more comprehensive testing of the BaseToolParser.

All changes appear to be focused on the tests directory, with no modifications to production code. If merged, this will contribute to a more robust and well-documented test suite.