Speed up Docker builds: .dockerignore + merge model download layers

.dockerignore excludes eval/, scripts/, tests/, data/, zips, and docs
so COPY steps only transfer what the app actually needs.

Merged two model-download RUN commands into one layer — fewer layers,
less cache store overhead.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

.dockerignore

@@ -0,0 +1,10 @@
+eval/
+scripts/
+tests/
+client/
+data/
+*.zip
+*.md
+*.toml
+*.ini
+Makefile

Dockerfile

@@ -11,8 +11,10 @@ COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Pre-download models so first request isn't slow on HF Spaces
-RUN python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('all-MiniLM-L6-v2')"
-RUN python -c "from sentence_transformers import CrossEncoder; CrossEncoder('vectara/hallucination_evaluation_model', trust_remote_code=True)"
+RUN python -c "\
+from sentence_transformers import SentenceTransformer, CrossEncoder; \
+SentenceTransformer('all-MiniLM-L6-v2'); \
+CrossEncoder('vectara/hallucination_evaluation_model', trust_remote_code=True)"
 
 COPY knowledge/ ./knowledge/
 COPY backend/   ./backend/