feat: add multi-modal attachment propagation to all worker agents thr…

[bittoby]

Closes: #956

Description

This PR enables all worker agents (Developer, Browser, Document, Social Media) to receive multi-modal information (image attachments) through task decomposition, not just the Multi-Modal Agent.

Problem: Previously, when users attached images to complex tasks, only the Multi-Modal Agent could access them. Worker agents received image file paths in additional_info but couldn't use them, causing subtasks requiring image analysis to fail.

Solution:

• Propagate parent task attachments to all subtasks during decomposition
• Include additional_info (image paths) in task assignment payloads
• Add comprehensive logging for attachment propagation tracking
• Add error handling to prevent failures when copying attachment metadata

Impact: Worker agents can now receive image context, enabling use cases like:

• Developer Agent analyzing UI screenshots to write code
• Browser Agent using reference images for web research
• Document Agent incorporating images into reports
• Social Media Agent analyzing images for content creation

What is the purpose of this pull request?

• Bug fix
• New Feature
• Documentation update
• Other

Note: This is both a bug fix (attachments weren't propagating) and a feature enabler (prepares for multi-modal toolkit integration in worker agents).

[bittoby]

@Wendong-Fan Could you please review this PR? thanks

[Wendong-Fan]

@Wendong-Fan Could you please review this PR? thanks

thanks @bittoby 's contribution! could @nitpicker55555 and @Zephyroam help checking this?