chore: clean model comment wording

feat: update image analysis prompts and training assets

chore: mock ML endpoints with Gemini

fix: improve image analysis reliability and error handling

feat: add digital ink detection to ELA heuristic

fix: remove CNN from build (OOM), inline tabular training, ELA heuristic fallback for image

fix: remove XGBoost use_label_encoder (removed in v2+), reduce CNN to 3k imgs/5 epochs for HF timeout, pre-download ResNet18

fix: load all HF dataset splits (validation+test), fix joblib CPU cores in container

fix: pin numpy<2.0 and transformers==4.40.2 for torch 2.2.2 compatibility

fix: remove saved_models from git (trained at Docker build time)

feat: hybrid CNN training with real HF cert images + fix Dockerfile offline vars bug

feat: scale training data to 25k rows + 5k CNN images

feat: migrate to HF Spaces — ResNet-18 CNN + BERT + DistilBERT + tabular ensemble

feat: optimize for 512MB memory (Render free tier)

FIX: optimize training models

UPADTE : created a endpoint to prevent sleep

CREATE : Trained the model