feat: support mixed text/image output

[YifeiDevs]

* denotes required fields

Priority*

• High: This PR needs to be merged first, before other tasks.
• Medium: This PR should be merged quickly to prevent conflicts due to common changes. (default)
• Low: This PR does not affect other tasks, so it can be merged later.

Purpose of the PR*

This PR addresses the issue where the plugin freezes and fails to render Base64 images returned by MCP. It aims to improve the rendering stability and capabilities for mixed content (text + images).

Changes*

• Enabled Base64 previews: Fixed the rendering logic to correctly handle Base64 image data.
• Batch uploads (images): Added support for processing batch images uploads.
• Context preservation: Ensured text context is preserved via placeholders during the rendering process.

How to check the feature

1. Use the plugin in an environment like Google AI Studio.
2. Trigger an action that returns a Base64 image via MCP.
3. Observe that the plugin no longer freezes.
4. Verify that both the image and the accompanying text context are rendered correctly.

Reference

Fixes #151

[github-actions]

Thank you for your contribution. We will check and reply to you as soon as possible.