update layout for demo 3 (#15975)

docs/version3.x/deployment/mcp_server.en.md

@@ -18,7 +18,7 @@ This project provides a lightweight [Model Context Protocol (MCP)](https://model
 ## Examples:
 The following showcases creative use cases built with PaddleOCR MCP server combined with other tools:
 
-### Demo 1: 
+### Demo 1
 In Claude for Desktop, extract handwritten content from images and save to note-taking software Notion. The PaddleOCR MCP server extracts text, formulas and other information from images while preserving document structure.
 <div align="center">
   <img width="65%" src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/paddleocr/mcp_demo/note_to_notion.gif" alt="note_to_notion">
@@ -29,7 +29,7 @@ In Claude for Desktop, extract handwritten content from images and save to note-
 
 ---
 
-### Demo 2: 
+### Demo 2
 In VSCode, convert handwritten ideas or pseudocode into runnable Python scripts that comply with project coding standards with one click, and upload them to GitHub repositories. The PaddleOCR MCP server extracts explicitly handwritten code from images for subsequent processing.
 
 <div align="center">
@@ -40,15 +40,17 @@ In VSCode, convert handwritten ideas or pseudocode into runnable Python scripts
 
 ---
 
-### Demo 3: 
+### Demo 3
 In Claude for Desktop, convert PDF documents or images containing complex tables, formulas, handwritten text and other content into locally editable files.
 
-- Convert complex PDF documents with tables and watermarks to editable doc/Word format:
+#### Demo 3.1
+Convert complex PDF documents with tables and watermarks to editable doc/Word format:
 <div align="center">
   <img width="70%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/paddleocr/mcp_demo/pdf_to_file.gif" alt="pdf_to_file">
 </div>
 
-- Convert images containing formulas and tables to editable csv/Excel format:
+#### Demo 3.2
+Convert images containing formulas and tables to editable csv/Excel format:
 <div align="center">
   <img width="70%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/00136903a4d0b5f11bd978cb0ef5d3c44f3aa5e9/images/paddleocr/mcp_demo/table_to_excel1.png" alt="table_to_excel1">
   <img width="50%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/00136903a4d0b5f11bd978cb0ef5d3c44f3aa5e9/images/paddleocr/mcp_demo/table_to_excel2.png" alt="table_to_excel2">

docs/version3.x/deployment/mcp_server.md

@@ -18,7 +18,7 @@
 ## 示例：
 以下展示了使用 PaddleOCR MCP 服务器结合其他工具搭建的创意案例：
 
-### Demo 1：
+### Demo 1
 在 Claude for Desktop 中，提取图像中的手写内容，并存到笔记软件 Notion。PaddleOCR MCP 服务器从图像中提取了文字、公式等信息，并保留了文档的结构。
 <div align="center">
   <img width="65%" src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/paddleocr/mcp_demo/note_to_notion.gif" alt="note_to_notion">
@@ -29,7 +29,7 @@
 
 ---
 
-### Demo 2：
+### Demo 2
 在 VSCode 中，根据手写思路或伪代码一键转换为可运行并符合项目代码风格规范的 Python 脚本，并将其上传到 GitHub 仓库中。PaddleOCR MCP 服务器从图像中准确提取手写代码供后续步骤使用。
 
 <div align="center">
@@ -40,15 +40,17 @@
 
 ---
 
-### Demo 3：
+### Demo 3
 在 Claude for Desktop 中，将含有复杂表格、公式、手写文字等内容的 PDF 文档或图片转存为本地可编辑文件。
 
-- 含表格水印复杂文档PDF 转为 doc/Word 可编辑格式：
+#### Demo 3.1
+含表格水印复杂文档PDF 转为 doc/Word 可编辑格式：
 <div align="center">
   <img width="70%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/main/images/paddleocr/mcp_demo/pdf_to_file.gif" alt="pdf_to_file">
 </div>
 
-- 含公式表格图片转为 csv/Excel 可编辑格式：
+#### Demo 3.2
+含公式表格图片转为 csv/Excel 可编辑格式：
 <div align="center">
   <img width="70%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/00136903a4d0b5f11bd978cb0ef5d3c44f3aa5e9/images/paddleocr/mcp_demo/table_to_excel1.png" alt="table_to_excel1">
   <img width="50%" img src="https://raw.githubusercontent.com/cuicheng01/PaddleX_doc_images/00136903a4d0b5f11bd978cb0ef5d3c44f3aa5e9/images/paddleocr/mcp_demo/table_to_excel2.png" alt="table_to_excel2">