fix: handle OpenAI model responses with tool calls and no other assistant content

[clareliguori]

Description

The model openai.gpt-oss-120b can sometimes respond with only toolUse blocks and no text content. When this happens, the OpenAI formatter creates an assistant message with an empty content array:

{'role': 'assistant', 'content': [], 'tool_calls': [...]}

The next request to Bedrock's Mantle OpenAI-compliant endpoint then fails with validation errors:

BadRequestError: 1 validation error for Message
content.0
  Input should be a valid dictionary or instance of Content

This change conditionally includes the content key in the assistant message only when formatted_contents is non-empty.

I also updated LiteLLMModel since it has the same logic.

Type of Change

Bug fix

Testing

How have you tested the change? Verify that the changes do not break functionality or introduce warnings in consuming repositories: agents-docs, agents-tools, agents-cli

• I ran hatch run prepare

Checklist

• I have read the CONTRIBUTING document
• I have added any necessary tests that prove my fix is effective or my feature works
• I have updated the documentation accordingly
• I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
• My changes generate no new warnings
• Any dependent changes have been merged and published

---

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!