feat: Add 176K real DrugBank DDI samples with drug names

- Downloaded 191K DDI pairs from TDC DrugBank
- Fetched 1,634 drug names from PubChem API (96% hit rate)
- Created complete training dataset with:
  - Real drug names (not just IDs)
  - 86 interaction type descriptions
  - Severity labels (minor/moderate/major/contraindicated)
- Bundled 34MB data file in Docker image
- Handler loads real data instead of curated samples

components/runpod_trainer/Dockerfile

@@ -8,11 +8,13 @@ COPY requirements.txt /app/requirements.txt
 # Install dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy handler
+# Copy handler and data
 COPY handler.py /app/handler.py
+COPY data/ /app/data/
 
 # Set environment variables
 ENV PYTHONUNBUFFERED=1
 ENV HF_HOME=/tmp/huggingface
+ENV DDI_DATA_PATH=/app/data/drugbank_ddi_complete.jsonl
 
 CMD ["python", "-u", "handler.py"]