chore: add .gitignore, remove PDFs/models/zips from history

.gitignore

@@ -1,41 +1,76 @@
-# Python
+# === Python ===
 __pycache__/
 *.py[cod]
 *.pyo
 *.egg-info/
+*.egg
 dist/
 build/
-*.spec
+*.whl
 
-# Environnement virtuel
+# === Virtual environments ===
 .venv/
 venv/
+venv_*/
 env/
 
-# IDE
+# === ML Models & Data ===
+*.pt
+*.pth
+*.onnx
+*.bin
+*.safetensors
+*.h5
+*.hdf5
+*.pkl
+*.pickle
+*.npy
+*.npz
+*.faiss
+models/
+*.tar.gz
+*.zip
+
+# === Documents & Media ===
+*.pdf
+*.docx
+*.xlsx
+*.csv
+*.png
+*.jpg
+*.jpeg
+*.gif
+*.mp3
+*.wav
+*.mp4
+
+# === IDE ===
 .idea/
 .vscode/
 *.swp
 *.swo
+*~
 
-# Modeles NER (volumineux, telecharges automatiquement)
-models/
-
-# PDF de test et resultats
-pdf_natif/
-pseudonymise/
-
-# Archives
-*.zip
-
-# Nuitka build
-*.build/
-*.dist/
-*.onefile-build/
-
-# OS
+# === OS ===
 .DS_Store
 Thumbs.db
+.~lock.*
 
-# Divers
-test-mini.js
+# === Secrets ===
+.env
+*.env
+credentials.json
+token.pickle
+
+# === Logs & Cache ===
+*.log
+logs/
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+htmlcov/
+.coverage
+
+# === Backups ===
+*_backup_*
+backups/