[michalmaj]

Summary

Add local statement and branch coverage reporting without introducing a percentage gate.

Why

Coverage can reveal source code that tests never execute, but a high percentage does not prove that assertions or requirements are correct. The guide should teach coverage as diagnostic evidence rather than a score to maximize.

Changes

• add pytest-cov as a locked development dependency
• configure branch measurement and missing-line reporting in pyproject.toml
• add a focused coverage chapter and connect it to the main learning path
• update configuration, glossary, common questions, and roadmap references

Type of change

Select the most relevant option.

• Documentation
• Project configuration
• New feature
• Tests
• CI/CD
• Refactoring
• Bug fix

Checklist

• The change is small and focused
• The pull request title is clear
• I updated documentation if needed
• I added or updated tests if needed
• I ran local checks if applicable

Notes

The project intentionally does not configure fail_under. The current report
is 100% statement and branch coverage because the example package is very small;
that result is not treated as a quality guarantee.

This is the bottom pull request in the coverage stack. The follow-up 8B pull
request adds coverage reporting to CI.

Local checks passed:

• uv run pytest --cov=text_toolkit --cov-report=term-missing
• uv run mypy
• uv run ruff check .
• uv run ruff format --check .
• uv build --no-sources
• git diff --check