PR_BODY_86.md

This PR implements consistent, structured logging across the codebase and satisfies ticket #86.

Summary

• Central utility autorepro/utils/logging.py with JSON and key=value text formatters.
• CLI configured via configure_logging(); honors AUTOREPRO_LOG_FORMAT=json.
• Decorators enriched with structured context (operation, args, result, duration_s).
• Replaced non-CLI print() with logging in rules and GitHub integrations.
• Docs updated with logging guidance.

Acceptance Criteria

• All modules use consistent logging patterns.
• No print() statements except CLI user output.
• Appropriate log levels used (DEBUG/INFO/WARNING/ERROR).
• Context included across messages; structured format supported (JSON).

Changes

• add: autorepro/utils/logging.py
• refactor: autorepro/cli.py logging setup
• refactor: autorepro/utils/decorators.py adds structured context
• refactor: autorepro/rules.py plugin load error handling uses logging
• refactor: autorepro/utils/github_api.py, autorepro/io/github.py dry-run prints -> INFO logs
• docs: CONTRIBUTING.md logging section

Usage

• Default text logs (stderr): key=value with timestamps.
• Structured logs: set AUTOREPRO_LOG_FORMAT=json in environment.
• Add context: logging.getLogger("autorepro").info("msg", extra={"operation": "plan"})

Closes #86