update custom llama cpp support

some experimental

Add LangSearch web search tool for small model knowledge augmentation

Copy llama.cpp shared libraries into runtime image

Speed up build (ccache) and fix /data permissions: create appuser and run downloader/server as appuser

Use Python venv in runtime to avoid PEP 668 pip install error

Fix llama.cpp math compiler flags

Fix llama.cpp Debian slim build deps

Add optimized llama.cpp CPU inference backend