RAG 🇯🇲

[vorburger]

FIRST finish #1731 (to be able to compare it...), and THEN make (something like) this RAG 🇯🇲 possible:

./enola rag load --in="docs/**" --rag=rag.json&type=langchain4j

./enola ai --llm="google://?model=gemini-2.5-flash-lite" --rag=rag.json&type=langchain4j  --prompt="What is Enola?"

with:

• https://github.com/enola-dev/enola/tree/main/java/dev/enola/ai/langchain4j/rag
• https://docs.enola.dev/specs/aiuri/#rag TBD
• https://docs.langchain4j.dev/tutorials/rag
• https://docs.langchain4j.dev/integrations/embedding-stores/in-memory