Production-ready CPET analysis

CPET analysis updates

obseed has updated its cardiopulmonary exercise testing (CPET) workflow with standardized oxidation models, automated ventilatory threshold detection, and persistent breath-by-breath data storage.

Fat oxidation and curve modeling

• Replaced legacy Cosmed fat-energy output with Frayn stoichiometric fat oxidation measured in grams per hour.
• Added physiologically constrained SIN and Gaussian curve models based on the Cheneviere and Amaro-Gahete approaches.

Threshold detection and analyst controls

• Added automatic VT1 detection using the V-slope method and VT2 detection from VE/VCO2, complete with confidence scoring and editable analyst controls.

Quality scoring and VO2max validation

• Added a VO2max validation checklist using RER, plateau, and heart-rate criteria.
• Introduced a composite analysis-quality score from 0 to 100 with a traffic-light breakdown.

Data persistence via parser API

• Persisted validation results and breath-by-breath data through the parser API, enabling analysis review beyond the Streamlit session.