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>

web/next.config.js

@@ -32,6 +32,10 @@ const nextConfig = {
         source: '/generated_images/:path*',
         destination: `${backendUrl}/generated_images/:path*`,
       },
+      {
+        source: '/images/:path*',
+        destination: `${backendUrl}/images/:path*`,
+      },
     ];
   },
   // 禁用静态生成，避免 SSR 时使用浏览器 API 的错误