[Fix] Images rejected as unsupported when using DeepSeek V4 Pro or V4 Flash

[zoomote]

Opened on behalf of Toray Altas. View the task or mention @zoomote for follow-up asks.

Related GitHub Issue

Closes: #964

Description

DeepSeek V4 Pro and V4 Flash were marked supportsImages: false in packages/types/src/providers/deepseek.ts, so image attachments were blocked with unsupported_model even though the DeepSeek API accepts image inputs. This PR flips supportsImages to true for deepseek-v4-pro and deepseek-v4-flash only; the legacy deepseek-chat and deepseek-reasoner aliases keep supportsImages: false as they do not support vision and are being retired. The image transport path already handles these models: convertToR1Format converts base64 image blocks to OpenAI image_url content parts, so no pipeline changes were needed. Regression assertions were added to deepseek.spec.ts covering supportsImages for both V4 models (the legacy aliases were already asserted as false).

Test Procedure

• pnpm test passes (10/10 turbo tasks), including the updated src/api/providers/__tests__/deepseek.spec.ts assertions that both V4 models report supportsImages: true and both legacy aliases still report false.
• Pre-commit hooks (lint + check-types across all packages) passed.
• Reviewers can verify by selecting deepseek-v4-pro or deepseek-v4-flash and attaching an image; it should no longer be rejected with unsupported_model.

Pre-Submission Checklist

• Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
• Scope: My changes are focused on the linked issue (one major feature/fix per PR).
• Self-Review: I have performed a thorough self-review of my code.
• Testing: New and/or updated tests have been added to cover my changes (if applicable).
• Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
• Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

Not applicable — this is a model metadata flag change with no UI rendering change.

Documentation Updates

• Yes, documentation updates are required. A paired docs PR in Zoo-Code-Org/Zoo-Code-Docs adds an image-input note to the DeepSeek provider page (see Related PRs).

Additional Notes

Related PRs

• Docs: [Docs] Document image input support for DeepSeek V4 models Zoo-Code-Docs#15

Linked work items

Closes #964

Get in Touch

Available via this PR's comments for any reviewer questions.

[zoomote]

No code issues found. See task

_{Reviewed 6fa4a70}

[zoomote]

Closing as duplicate of #963, which was already open for the same fix. The paired docs PR (Zoo-Code-Docs#15) remains open.