fix dockerfile, pyproject.toml

Dockerfile

@@ -1,14 +1,19 @@
 FROM python:3.12-slim
 
+# Install build dependencies for llama-cpp-python
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    gcc \
+    g++ \
+    && rm -rf /var/lib/apt/lists/*
+
 RUN pip install --no-cache-dir uv
 
 WORKDIR /app
 
 COPY pyproject.toml /app/pyproject.toml
 
-# Install dependencies with binary wheels (avoid compilation)
-ENV CMAKE_ARGS="-DLLAMA_BLAS=OFF -DLLAMA_AVX2=OFF" \
-    FORCE_CMAKE=0
+# Install dependencies (llama-cpp-python will compile but with minimal features)
+ENV CMAKE_ARGS="-DLLAMA_BLAS=OFF -DLLAMA_AVX2=OFF -DLLAMA_FMA=OFF -DLLAMA_F16C=OFF"
 RUN uv pip install --system -r pyproject.toml
 
 COPY . /app