install flash attention

Dockerfile

@@ -45,10 +45,10 @@ COPY --chown=user requirements.txt .
 RUN --mount=type=cache,target=$UV_CACHE_DIR,uid=1000,gid=1000 \
     /uv pip install -r requirements.txt
 
-# Conditionally install flash-attn if CUDA is available
+# Try to install flash-attn, but don't fail the build if it errors
 RUN --mount=type=cache,target=$UV_CACHE_DIR,uid=1000,gid=1000 \
-    python -c "import torch; exit(0 if torch.cuda.is_available() else 1)" && \
-    /uv pip install flash-attn --no-build-isolation || echo "CUDA not available, skipping flash-attn installation"
+    echo "Attempting to install flash-attn..." && \
+    (/uv pip install flash-attn --no-build-isolation || echo "Failed, continuing build...")
 
 # Copy application code
 COPY --chown=user . .