fix: add shap to requirements for deep explanation endpoint

fix: f-string backslash syntax error in generate_explanation_text

refactor: clean up inference.py, remove inline comments, extract _patch_xlnet_configs

fix: use slow tokenizer for roberta to avoid tokenizer.json format incompatibility

fix: remove train/evaluate imports from __init__, pin tokenizers==0.19.1

fix: pin transformers==4.44 and fastapi==0.111 to avoid breaking changes

fix: torch 2.4+cpu and numpy<2 to fix NumPy 2.x incompatibility

fix: minimal Dockerfile and requirements for fast build

fix: use CPU-only torch to cut build time from 10min to 2min

fix: default MODEL_NAMES to fine-tuned HF repos, fix .gitignore

chore: force rebuild

fix: remove build-time model download, load from HF Hub at runtime

fix: slim requirements.txt and fix Docker build deps

feat: Phase 1 enhancements - ensemble endpoint, history API, rate limiting, storage monitoring

fix: smart MODELS_DIR path resolution for Docker

fix: use hf download instead of deprecated huggingface-cli

fix: remove deprecated use_cache from xlnet config

chore: clean up preload log message

fix: patch xlnet tokenizer_config at runtime

fix: rebuild to pick up fixed xlnet tokenizer_config

fix: preload models synchronously

fix: sync latest inference and api fixes

fix: Python 3.9 compat list[dict] type hints

fix: preload all models at startup

chore: bust pip cache to force full rebuild

fix: evict model cache to prevent OOM

chore: bust docker cache to force src/ rebuild

fix: remove syntax error in xlnet inference

fix: xlnet list fields in _token_importance

fix: xlnet tokenizer list fields

fix: distilbert token_type_ids and xlnet inference errors

Remove .env.example from Dockerfile

Fix colorFrom metadata

Initial deployment

initial commit