feat(stt): support auto language detection for mixed Chinese-English input

- Flutter: language='auto' omits the language field → backend receives none
- Backend: no language field → passes undefined to STT service
- STT service: language=undefined → omits language param from Whisper request
- Whisper auto-detects language per utterance when no hint is provided

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

it0_app/lib/features/chat/data/datasources/chat_remote_datasource.dart

@@ -154,17 +154,19 @@ class ChatRemoteDatasource {
   /// Backend performs STT, interrupts any running task if needed, and
   /// Transcribe audio to text (STT only, does NOT send to agent).
   /// Returns the transcript string.
+  /// Pass language='auto' to let Whisper auto-detect (best for mixed-language).
   Future<String> transcribeAudio({
     required String audioPath,
     String language = 'zh',
   }) async {
-    final formData = FormData.fromMap({
+    final fields = <String, dynamic>{
       'audio': await MultipartFile.fromFile(
         audioPath,
         filename: audioPath.split('/').last,
       ),
-      'language': language,
+    };
-    });
+    if (language != 'auto') fields['language'] = language;
+    final formData = FormData.fromMap(fields);
     final response = await _dio.post(
       '${ApiEndpoints.agent}/transcribe',
       data: formData,

packages/services/agent-service/src/infrastructure/stt/openai-stt.service.ts

@@ -34,7 +34,7 @@ export class OpenAISttService {
   async transcribe(
     audioBuffer: Buffer,
     filename: string,
-    language = 'zh',
+    language?: string,
   ): Promise<string> {
     const url = `${this.baseUrl}/v1/audio/transcriptions`;
     this.logger.log(
@@ -54,7 +54,7 @@ export class OpenAISttService {
     };
 
     appendField('model', 'whisper-1');
-    appendField('language', language);
+    if (language) appendField('language', language); // omit → Whisper auto-detects
     appendField('response_format', 'json');
 
     // File field

packages/services/agent-service/src/interfaces/rest/controllers/agent.controller.ts

@@ -421,10 +421,11 @@ export class AgentController {
     if (!file?.buffer?.length) {
       throw new BadRequestException('audio file is required');
     }
+    // language=undefined → Whisper auto-detects (best for mixed-language input)
     const text = await this.sttService.transcribe(
       file.buffer,
       file.originalname || 'audio.m4a',
-      language ?? 'zh',
+      language || undefined,
     );
     return { text: text?.trim() ?? '' };
   }