Feature Request: Add native image_input support for model deepseek-v4-flash

[xylxrp]

Is your feature request related to a problem? Please describe.
I am trying to enable image input capability for the deepseek-v4-flash model (DeepSeek V4 Flash) via config.yaml. Currently, the 📎 image upload button does not appear in the Continue chat panel for this model, even though the model itself supports native multimodal understanding.

Describe the solution you'd like
Please add deepseek-v4-flash to Continue's internal capability auto-detection list with image_input: true. DeepSeek V4 Flash officially supports native image recognition/understanding.

Additional context
My current configuration in config.yaml:

  - name: DeepSeek V4 Flash
    provider: openai
    model: deepseek-v4-flash
    apiBase: https://api.deepseek.com/v1
    capabilities:
      - image_input
      - tool_use

The button appears fine for deepseek-v4-pro or gpt-4o, just not for deepseek-v4-flash.

Environment:

• VS Code Version: 1.120.0
• OS: Windows 10