feat: multimodal RAG with PDF image extraction and display

Extract images from PDFs using pymupdf, generate descriptions via
Qwen3-VL-8B, store in ChromaDB alongside text chunks, and render
images in chat answers. Includes image proxy rewrite, force re-process
endpoint, and VLM API timeout.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

backend/pyproject.toml

@@ -42,6 +42,7 @@ dependencies = [
     "duckduckgo-search>=6.0.0",
     "docx2txt>=0.9",
     "pypdf>=6.12.0",
+    "pymupdf>=1.27.2.3",
 ]
 
 [project.optional-dependencies]