Add rag module: Spring AI 2.0 RAG with pgvector, chunking, reranking and a faithfulness check

Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01B38FGKKam5SCGgwgduVAh3

rag/init.sql

@@ -0,0 +1,20 @@
+-- The schema the application expects. docker-compose.yml runs this on first start, and
+-- scripts/pg-up.sh runs it against a local PostgreSQL. initialize-schema is false in
+-- application.yml, so the application never creates or alters any of this itself.
+
+CREATE EXTENSION IF NOT EXISTS vector;
+
+CREATE SCHEMA IF NOT EXISTS rag;
+
+-- metadata is json, the type Spring AI's PgVectorStore writes and its filter queries cast from.
+-- 1536 is the size of a text-embedding-3-small vector; change it together with the model.
+CREATE TABLE IF NOT EXISTS rag.document_chunks (
+    id        uuid PRIMARY KEY DEFAULT gen_random_uuid(),
+    content   text,
+    metadata  json,
+    embedding vector(1536)
+);
+
+-- HNSW with cosine distance, matching index-type and distance-type in application.yml.
+CREATE INDEX IF NOT EXISTS document_chunks_embedding_idx
+    ON rag.document_chunks USING hnsw (embedding vector_cosine_ops);