discussion: expanding the use-case of llama.cpp - embedded LLM toolchain

[jon-chuang]

For instance, should llama.cpp:

1. support an embedded vector-similarity knowledge base?
2. support other models for multimodality (similar to GPT4). (See e.g. CLIP-based)

It doesn't have be part of the compilation of main, it could be in a llama.embed toolchain