copy rpc-server binary to runtime container

toolboxes/Dockerfile.rocm-7rc

@@ -74,11 +74,10 @@ RUN cmake -S . -B build \
  && cmake --build build --config Release -- -j$(nproc) \
  && cmake --install build --config Release
 
-# keep bin; drop headers/docs/static libs; drop source tree
+# keep bin; drop headers/docs/static libs (retain llama.cpp for rpc binaries)
 RUN find /opt/rocm-7.0 -type f -name '*.a' -delete \
  && rm -rf /opt/rocm-7.0/include /opt/rocm-7.0/share \
-           /opt/rocm-7.0/llvm/include /opt/rocm-7.0/llvm/share \
- && rm -rf /opt/llama.cpp
+          /opt/rocm-7.0/llvm/include /opt/rocm-7.0/llvm/share
 
 # runtime
 FROM registry.fedoraproject.org/fedora-minimal:43
@@ -89,6 +88,7 @@ RUN microdnf -y --nodocs --setopt=install_weak_deps=0 install \
 
 COPY --from=builder /opt/rocm-7.0 /opt/rocm-7.0
 COPY --from=builder /usr/local/ /usr/local/
+COPY --from=builder /opt/llama.cpp/build/bin/rpc-* /usr/local/bin/
 
 COPY gguf-vram-estimator.py /usr/local/bin/
 RUN chmod +x /usr/local/bin/gguf-vram-estimator.py