fix(mcp): update code example highlight line ranges (#36)

Co-authored-by: Chai Landau <[email protected]>

Author: tuliren

mcp/intro.mdx

@@ -172,31 +172,31 @@ A tool is a function exposed by the server that an AI agent can call using `JSON
 
 Resources are named, read-only data references addressable via URIs. Resources are used to load contextual read-only data into the MCP session.
 
-```python ! mcp-server.py focus=19:24
+```python ! mcp-server.py focus=19:25
 !from ./assets/intro/mcp.py.txt
 ```
 
 ## !!steps prompts
 
 Prompts are predefined message templates that servers expose to clients. When invoked with arguments, they return a list of messages (`messages[]`) that the client sends to the LLM to perform a specific task.
 
-```python ! mcp-server.py focus=26:31
+```python ! mcp-server.py focus=27:32
 !from ./assets/intro/mcp.py.txt
 ```
 
 ## !!steps roots
 
 Roots are a set of scoped access boundaries provided by the client at handshake. Roots allow the server to limit the visibility of tools, resources, or data based on a given namespace.
 
-```python ! mcp-server.py focus=33:38
+```python ! mcp-server.py focus=34:40
 !from ./assets/intro/mcp.py.txt
 ```
 
 ## !!steps transports
 
 A transport defines the communication method the server uses to handle client connections. MCP supports multiple transports, including `stdio` for local execution and `http` for web-based deployment.
 
-```python ! mcp-server.py focus=40:53
+```python ! mcp-server.py focus=45:58
 !from ./assets/intro/mcp.py.txt
 ```