feat: replace server-side RAG with MCP retrieval primitives

- Remove Phase 4 RAG pipeline in favor of retrieval-only architecture
- Add FastMCP server exposing search, get_chunk, list_libraries tools
- Mount MCP endpoints (streamable HTTP + SSE) via Starlette in ASGI config
- Update README to clarify Mnemosyne is a retrieval engine, not RAG
- Let calling LLMs drive synthesis and iterative retrieval themselves

mnemosyne/mcp_server/context.py

@@ -0,0 +1,19 @@
+"""Helpers for accessing the request-scoped MCP user/token from inside tools."""
+
+from __future__ import annotations
+
+from fastmcp.server.context import Context
+
+from .auth import STATE_KEY_TOKEN, STATE_KEY_USER
+
+
+async def get_mcp_user(ctx: Context | None):
+    if ctx is None:
+        return None
+    return await ctx.get_state(STATE_KEY_USER)
+
+
+async def get_mcp_token(ctx: Context | None):
+    if ctx is None:
+        return None
+    return await ctx.get_state(STATE_KEY_TOKEN)