[Tools] PID Feature Extractor task for particle identification in O2Physics

[forynski]

This pull request introduces a new O2Physics analysis task, PID Feature Extractor, designed to extract particle identification (PID) features from ALICE AO2D data. It integrates well with the O2Physics framework and supports both real and simulated data.

Features and Functionality:

• Extracts 39 PID-related features per track, including kinematic variables, TPC and TOF detector responses, Bayesian combined PID probabilities, and MC truth where available.
• Combines TPC dE/dx and TOF timing information using a Bayesian approach with configurable priors to output PID probabilities for pion, kaon, proton, and electron hypotheses.
• Automatic fallback to ALICE Conditions Database (CCDB) for PID calibrations when not available in AO2D files, using collision timestamps for correct calibration periods.
• Outputs feature data in both ROOT TTree and CSV formats.
• Includes quality control histograms for track and detector response monitoring.
• Configuration is managed via a JSON file (myConfigExtractor.json), allowing easy adjustment of selection criteria, output formats, and CCDB server endpoint.
• Comes with an execution script (run.sh) for streamlined running inside the O2Physics environment.
• Fully documented with README.md detailing installation, usage, configuration, and machine learning integration.

Repository Changes:

• Added new task folder: Tools/PIDFeatureExtractor/ with source files, configuration, scripts, and documentation.
• Updated Tools/CMakeLists.txt to include PIDFeatureExtractor in the build.

Testing and Validation:

• Successfully built and integrated within O2Physics build system.
• Validated feature extraction on AO2D test datasets with proper output to ROOT and CSV.
• Ensured CCDB integration works with fallback mechanism.
• JSON configuration file validated and documented.
• Script permissions set correctly for execution.

Impact:

• Enhances O2Physics capabilities by providing ready-to-use, ML-friendly PID feature extraction.
• Useful for particle identification model training, performance studies, and physics analyses.
• Well encapsulated, minimally invasive changes limited to the Tools directory.
• Sets foundation for future PID-related enhancements and expansions.

[github-actions]

O2 linter results: ❌ 0 errors, ⚠️ 1 warnings, 🔕 0 disabled

[github-actions]

This PR has not been updated in the last 30 days. Is it still needed? Unless further action is taken, it will be closed in 5 days.

[forynski]

Is there any chance for a reviewer to look into this? Thanks in advance @alibuild @jgrosseo @iarsene @ktf @ddobrigk