[Agent V2] Missing vision toggle and model-capability awareness in configure UI

### Self Checks

- [x] I have read the [Contributing Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) and [Language Policy](https://github.com/langgenius/dify/issues/1542).
- [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones.
- [x] I confirm that I am using English to submit this report, otherwise it will be closed.
- [x] 【中文用户 & Non English User】请使用英语提交，否则会被关闭 ：）
- [x] Please do not modify this template :) and fill in all the required fields.

### Dify version

1.16.1

### Cloud or Self Hosted

Self Hosted (Docker)

### Steps to reproduce

1。Configure the vision model in the agent.
2.Attach images when asking questions

### ✔️ Expected Behavior

The Agent V2 configure UI should check the bound model's feature flags (specifically ModelFeature.VISION) when image uploads are configured.
Ideally, a vision-specific toggle or at minimum a warning should be shown when image uploads are enabled but the model does not advertise vision support.
Vision-specific settings (image resolution detail) should be accessible from the Agent V2 configure UI, mirroring the legacy ConfigVision component.

### ❌ Actual Behavior

The Agent V2 "Chat Features" panel shows a generic File Upload card that bundles document, image, audio, and video together under allowed_file_types [[2]](https://github.com/langgenius/dify/blob/8bdf702f737e31bc1f9e75def597e639f9b01f8c/web/features/agent-v2/agent-detail/configure/components/preview/chat-features-panel.tsx#L44-L44).
There is no vision-capability check anywhere in the Agent V2 configure context — the UI never verifies whether the bound model supports vision [[3]](https://github.com/langgenius/dify/blob/8bdf702f737e31bc1f9e75def597e639f9b01f8c/web/features/agent-v2/agent-detail/configure/components/preview/chat-features-panel.tsx) [[4]](https://github.com/langgenius/dify/issues/40874#issuecomment-5364539578).
There are no vision-specific settings (e.g., image resolution detail) accessible from Agent V2.
Users can enable image uploads without any indication that the selected model may not support vision input.

Start with web/features/agent-v2/agent-detail/configure/components/preview/chat-features-panel.tsx and trace the Agent V2 configure context for the bound model's feature flags. Compare the generic File Upload behavior with the legacy ConfigVision component, including image resolution detail. Done means image uploads reflect ModelFeature.VISION support and vision-specific settings or a clear warning are available in the Agent V2 configure UI.

Written by the indexing model from the issue text.