MCP tools should not return the full `CallToolResult`

[samuelcolvin]

Initial Checks

• I confirm that I'm using the latest version of Pydantic AI
• I confirm that I searched for my issue in https://github.com/pydantic/pydantic-ai/issues before opening this issue

Description

• if all parts are text, I think we should return it as concatenated text, perhaps '\n'.join(c.text for c in call.context)
• if they're all images, we should presumably return images images are multimodal chunks
• not sure what to do with embedded resources

The most common case is just to return a single chunk of text.

Example Code

Python, Pydantic AI & LLM client version

-

[samuelcolvin]

Here's what we're returning now, which is very excessive:

[DouweM]

Fixed in #1618