マルチモーダル対応モデルをわかりやすくしたい

[tegnike]

@claude

• src/features/constants/aiModels.ts で規定されているマルチモーダルモデルだが、AIサービスの設定画面では、どれが対象モデルか一目でわからない
• ユーザ目線でこれを一目でわかるようにする

[tegnike]

@claude

こちら対応をお願いします。
ブランチは必ずdevelopから切ってください。

[claude]

Claude finished @tegnike's task in 5m 18s —— View job • claude/issue-412-20250602_225629 • Create PR ➔

---

Todo List:

• Read repository guidelines (CLAUDE.md)
• Analyze current aiModels.ts structure
• Examine AI settings UI components
• Implement visual indicators for multimodal models
• Test the implementation
• Commit changes and provide PR link

Implementation Details:

Added multimodal model definitions and visual indicators:

1. Enhanced src/features/constants/aiModels.ts:

   • Added multiModalModels constant defining which models support vision capabilities
   • Added isMultiModalModel() function to check if a specific model supports multimodal
   • Added getMultiModalModels() function to get multimodal models for a service
   • Included OpenAI realtime/audio models as multimodal-capable

2. Updated src/components/settings/modelProvider.tsx:

   • Added 📷 camera emoji indicator next to all multimodal-capable models
   • Updated all AI service model dropdowns: OpenAI, Anthropic, Google, Groq, Cohere, Mistral AI, Perplexity, Fireworks, DeepSeek
   • Included OpenAI realtime and audio model dropdowns

Result: Users can now easily identify which models support image input with the 📷 icon in all model selection dropdowns.

Changes Committed:

• ✅ Commit 50858e3: Added visual indicators for multimodal models
• 📁 Modified files: src/features/constants/aiModels.ts, src/components/settings/modelProvider.tsx