feat(audio): add native audio support to type system and model clients

- Add AudioSource interface and 'audio' variant to MessageContentPart union
- Update buildUserMessage() to create audio content parts from attachments
- Add attachmentToAudioSource(), hasAudio(), stripAudioParts() helpers
- Gemini: native audio via inlineData (same format as images)
- OpenAI/GitHub: native audio via input_audio content parts
- Anthropic/Bedrock: graceful fallback to transcript text
- Update getMessageTextWithTools() to handle audio blocks for local models

src/models/anthropic.ts

@@ -41,6 +41,13 @@ function toAnthropicContent(content: string | MessageContentPart[]): string | un
         },
       };
     }
+    // Audio — Anthropic doesn't support native audio input; use transcript fallback
+    if (part.type === 'audio') {
+      if (part.source.transcript) {
+        return { type: 'text', text: `[Voice message]: ${part.source.transcript}` };
+      }
+      return { type: 'text', text: '[Audio message received but no transcript available]' };
+    }
     return part;
   });
 }