feat: UI nav, onboarding, L2CS weights path + torch.load; trim dev files

Integrate L2CS-Net gaze estimation

MERGE EVERYTHING from feature/integration-cleaned :((((((((((((((((((((((((((((((((((((((((((((((( so tired

evaluation: channel ablation script + feature importance LOPO

refactor: single checkpoints/ dir, fix uvicorn cmd and hybrid config path

chore: MLP pipeline, evaluation updates, feature importance, confusion matrices

chore: integration cleanup — remove eye CNN, add threshold justification, fix pipeline

Put all the models together (expect UI)

integrated open/closed eye cnn in script

reorg folders, add MLP demo

feat: add optional eye model (YOLO/MobileNet) alongside geometry

feat: stage 2 head pose, eye behaviour, MAR/yawn, tighter focus; add torch deps

chore: add template layout and stage 1 face mesh demo

set up template with folder structure and training script