feat: Set up comprehensive Python testing infrastructure

[llbbl]

Set Up Python Testing Infrastructure

Summary

This PR establishes a comprehensive testing infrastructure for the MetaVec project, migrating from conda-based dependency management to Poetry and setting up a complete testing framework.

Changes Made

Package Management

• Added Poetry configuration in pyproject.toml with full dependency specification
• Migrated dependencies from environment.yml to Poetry format with Python 3.11 compatibility
• Updated Python versions and package versions to work with current environment

Testing Framework

• Configured pytest with comprehensive settings including:
  • Coverage reporting (80% threshold)
  • HTML and XML coverage reports
  • Custom test markers (unit, integration, slow)
  • Strict configuration and test discovery patterns

Directory Structure

• Created testing directories:
  • tests/ - Root testing directory
  • tests/unit/ - Unit tests
  • tests/integration/ - Integration tests
  • All with proper __init__.py files

Testing Utilities

• Comprehensive fixtures in tests/conftest.py:
  • Temporary directories for file-based tests
  • Sample embeddings and vocabulary for ML testing
  • Mock configurations and PyTorch device setup
  • Environment cleanup and seeding utilities

Development Workflow

• Poetry scripts for running tests:
  • poetry run test - Run all tests
  • poetry run tests - Alternative command
• Validation tests to verify infrastructure works correctly
• Updated .gitignore with testing artifacts and IDE files

Testing Instructions

Installation

poetry install

Running Tests

# Run all tests
poetry run test

# Run with verbose output
poetry run pytest -v

# Run only unit tests
poetry run pytest -m unit

# Run with coverage report
poetry run pytest --cov=MetaVec --cov-report=html

Coverage Reports

• Terminal: Displays missing coverage after test runs
• HTML: Generated in htmlcov/ directory
• XML: Generated as coverage.xml for CI integration

Notes

• Dependencies updated to Python 3.11 compatible versions
• Poetry lock file preserved in version control for reproducible builds
• No actual unit tests added - only testing infrastructure and validation
• Ready for development - developers can immediately start writing tests using the established patterns

Validation

All validation tests pass successfully:

• ✅ Basic pytest functionality
• ✅ NumPy and scientific computing stack
• ✅ Fixture system working correctly
• ✅ Test markers and categorization
• ✅ Temporary file handling
• ✅ Coverage reporting integration