feat: ASR语音转写管线 + 群聊身份混淆修复

- 新增ASR语音识别管线: QQ语音→下载音频→qwen3-asr-flash转录→注入用户消息
- 模型名称全部从models.json路由获取,无硬编码
- 修复群聊中AI将非管理员用户误称为管理员昵称(叶酱)的问题
  - 助手回复缓存时标注[回复 昵称 (UID)],防止对话历史中身份混淆
  - 群聊上下文指令改为肯定性表述,移除具体名称提及
- trace面板时间戳改为YYYY-MM-DD HH:MM:SS格式,耗时统一显示为秒
- 修复Go time.Duration纳秒值在前端显示问题(Duration/1e6转毫秒)
- 新增video_tool插件模板
- 优化OpenAI adapter reasoning_content处理

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
2026-05-31 16:46:47 +08:00
parent d112fdd540
commit a9c79d7887
16 changed files with 780 additions and 67 deletions
@@ -31,13 +31,14 @@ type UnifiedMessage struct {
BotUID string `json:"-"` // bot's own platform UID, set by router
}
// Attachment represents a file/image/voice attachment.
// Attachment represents a file/image/voice/video attachment.
type Attachment struct {
Type string `json:"type"` // "image", "voice", "file", "video"
URL string `json:"url,omitempty"`
FileName string `json:"file_name,omitempty"`
MimeType string `json:"mime_type,omitempty"`
Size int64 `json:"size,omitempty"`
Duration int `json:"duration,omitempty"` // video/voice duration in seconds
}
// UnifiedResponse is AI-Core's response converted to unified format.