Prepare for public git: .gitignore, .env, sanitized configs

- Add .gitignore excluding .env, __pycache__, test_files (copyrighted PDFs),
  session logs, marker/ submodule, and other generated/private files
- Add .env with local config (gitignored); expand .env.example with all
  supported environment variables
- docker-compose.yml: all hardcoded values now use default pattern
- Containerfile: install marker-pdf from GitHub instead of local submodule;
  fix AMD_COMPUTE default to false; sanitize default OLLAMA_HOST
- Remove marker/ submodule from git tracking (keep local checkout via .gitignore)
- Remove session transcript and copyrighted ENISA PDFs from git tracking
- Sanitize local IPs in README, kubernetes deployment, deploy-podman.sh
- Fix hardcoded paths in convert_both.py and compare_conversions.py
- Update requirements.txt to reference GitHub instead of local --editable

.env.example

@@ -7,22 +7,47 @@
 # podman --env, or docker-compose environment.
 
 # -----------------------------------
-# Ollama (OCR fallback)
+# Authentication
 # -----------------------------------
-OLLAMA_HOST=http://10.0.1.127:11434
-DEESEEK_OCR_MODEL=deepseek-ocr
+API_KEY=                           # API key for X-API-Key auth (empty = disabled)
+SELF_URL=http://localhost:8001      # Public URL for file download/upload links
 
 # -----------------------------------
-# AMD GPU (set TRUE on AMD/ROCm systems)
+# OCR Backend
 # -----------------------------------
-AMD_COMPUTE=false           # set to "true" to enable ROCm
-TORCH_DEVICE=               # override PyTorch device ("rocm", "cuda", "cpu")
-MODEL_DTYPE=float32         # "float32" or "bfloat16" (bfloat16 on capable GPUs)
+OCR_BACKEND=marker                 # "marker" (marker-pdf) or "deepseek" (deepseek-ocr via ollama)
+
+# -----------------------------------
+# DeepSeek OCR (via Ollama)
+# -----------------------------------
+DEEPSEEK_OLLAMA_HOST=http://localhost:11434
+DEEPSEEK_OCR_MODEL=deepseek-ocr:latest
+DEEPSEEK_OCR_PROMPT=<|grounding|>Free OCR.
+
+# -----------------------------------
+# Ollama (LLM correction)
+# -----------------------------------
+OLLAMA_HOST=http://localhost:11434
+DEESEEK_OCR_MODEL=deepseek-ocr
+
+USE_LLM=false                      # Use an LLM to improve conversion accuracy
+LLM_SERVICE=marker.services.ollama.OllamaService
+
+OPENAI_BASE_URL=                   # For OpenAI-compatible LLM API
+OPENAI_API_KEY=                    # API key for OpenAI-compatible service
+OPENAI_MODEL=                      # Model name for OpenAI-compatible service
+
+# -----------------------------------
+# AMD GPU (set to true on AMD/ROCm systems)
+# -----------------------------------
+AMD_COMPUTE=false                  # set to "true" to enable ROCm
+TORCH_DEVICE=                      # override PyTorch device ("rocm", "cuda", "cpu")
+MODEL_DTYPE=float32                # "float32" or "bfloat16" (bfloat16 on capable GPUs)
 
 # -----------------------------------
 # Application
 # -----------------------------------
-PORT=8000
+PORT=8001
 HOST=0.0.0.0
 FLASK_APP=app
 FLASK_DEBUG=0
@@ -30,13 +55,13 @@ FLASK_DEBUG=0
 # -----------------------------------
 # Gunicorn (production)
 # -----------------------------------
-GUNICORN_WORKERS=2
+GUNICORN_WORKERS=3
 GUNICORN_THREADS=4
-GUNICORN_TIMEOUT=300
+GUNICORN_TIMEOUT=600
 GUNICORN_LOGLEVEL=info
 
 # -----------------------------------
-# Marker
+# Paths
 # -----------------------------------
-MARKER_OUTPUT_DIR=/app/conversion_results
+RESULTS_DIR=/app/conversion_results
 UPLOAD_DIR=/app/uploads