docs: Clarify new_items output conditions for improved developer experience

[MuhammadHamidRaza]

This Pull Request focuses on improving the documentation for the new_items property within RunResult and RunResultStreaming. The update aims to provide a clearer, more organized, and user-friendly explanation of how different RunItem types are generated.

The current documentation lists all RunItem types, but does not explicitly explain that they are mutually exclusive outcomes of specific agent actions (e.g., a handoff will not also generate a tool call item). This can lead to confusion and wasted time for new users trying to understand the agent's output.

Key changes and benefits:

• Categorized Output for new_items: The RunItems are now grouped and described based on the specific agent action that triggers them (direct message, tool use, handoff, or internal reasoning).
• Improved Clarity: This structure makes it immediately clear what to expect in the new_items list for any given agent run, directly addressing a common point of confusion for beginners.
• Saves Developer Time: By clarifying the output conditions, this change will save significant debugging and exploration time for new developers, making the SDK more accessible and intuitive.

I believe this update is a crucial improvement for the developer onboarding process and will reduce unnecessary frustration.

[MuhammadHamidRaza]

Merge It because it's soo easy for new developer they can understand easily. Which thing i faced hard so it's my opinion how i can make it easy

[seratch]

Thanks, but this page already has links to the reference documents and I think they already have enough guidance. We'd like to avoid duplicate and repetitive documents, so let me close this one.