fix: preserve assistant messages with reasoning_content in sanitize pass

[renchonghan]

Link Bug Report #8482

修复 _sanitize_assistant_messages 上传含 reasoning_content 的 assistant 历史消息时，content 和 tool_calls 都为空导致的 API 返回400错误

背景：
现有逻辑在对 assistant 消息做空内容清理时，只要 reasoning_content content tool_calls 有一个不为空就会放行。
但 DeepSeek V4 系列、MiMo 系列等推理模型要求历史 assistant 消息中content 和 tool_calls 至少有一个不为空，否则 API 返回 400 错误。

改动：将有 reasoning_content 但无 content/tool_calls 的消息进行占位保留，设置 content="" 保证content 和 tool_calls 至少有一个不为空的API要求；
仅对三者全空的真正垃圾消息执行过滤，同时日志级别从 warning 降为 debug，减少噪音。

Modifications / 改动点

• astrbot/core/provider/sources/openai_source.py — _sanitize_assistant_messages() 方法内条件分支重构

• This is NOT a breaking change. / 这不是一个破坏性变更。

Screenshots or Test Results / 运行截图或测试结果

已在实际环境中验证，DeepSeek V4 系列模型的 tool_calls 和多轮推理均恢复正常。

---

Checklist / 检查清单

• 😊 If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc.
  / 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。

• 👀 My changes have been well-tested, and "Verification Steps" and "Screenshots" have been provided above.
  / 我的更改经过了良好的测试，并已在上方提供了“验证步骤”和“运行截图”。

• 🤓 I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in requirements.txt and pyproject.toml.
  / 我确保没有引入新依赖库，或者引入了新依赖库的同时将其添加到 requirements.txt 和 pyproject.toml 文件相应位置。

• 😮 My changes do not introduce malicious code.
  / 我的更改没有引入恶意代码。

Summary by Sourcery

Bug Fixes:

• Prevent API 400 errors by ensuring assistant messages with reasoning_content but no content or tool_calls are preserved with a placeholder content value.

[sourcery-ai]

Hey - I've left some high level feedback:

• The nested if inside if _is_empty(content) and not tool_calls: is a bit hard to follow; consider extracting the reasoning_content handling into a small helper or using early returns/continues to flatten the control flow for readability.
• The log message for discarded messages now mentions (无 context | tool_calls | reasoning_content) while the code checks content, tool_calls, and reasoning_content; aligning the terminology in the log with the actual field names would avoid confusion when debugging.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- The nested `if` inside `if _is_empty(content) and not tool_calls:` is a bit hard to follow; consider extracting the `reasoning_content` handling into a small helper or using early returns/continues to flatten the control flow for readability.
- The log message for discarded messages now mentions `(无 context | tool_calls | reasoning_content)` while the code checks `content`, `tool_calls`, and `reasoning_content`; aligning the terminology in the log with the actual field names would avoid confusion when debugging.

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[gemini-code-assist]

Code Review

This pull request updates the empty assistant message filtering logic in openai_source.py to properly handle reasoning models. When a message contains reasoning_content but lacks content and tool_calls, it now sets msg["content"] to an empty string to satisfy API validation instead of discarding it. The reviewer noted a minor typo in the new debug log message, where "context" was used instead of "content", and provided a code suggestion to correct it.