feat(ai-gateway): prepare Vercel model provider options

[chrarnoldus]

Summary

• add Vercel provider-option generation for Anthropic and OpenAI model variants
• translate chat and Responses API verbosity into Anthropic effort options
• prepare Anthropic speed: fast and OpenAI reasoningMode: pro options for future model routing support
• centralize model variant detection in shared provider helpers and extend the Vercel provider-option types
• add focused coverage for provider-option generation

This PR intentionally does not map Opus fast or GPT Pro model IDs to base Vercel model IDs yet. That routing can be enabled separately once the model-name behavior is ready to ship.

Verification

• pnpm --filter web test -- --runInBand src/lib/ai-gateway/providers/vercel/index.test.ts src/lib/ai-gateway/providers/vercel/mapModelIdToVercel.test.ts
• pnpm --filter web typecheck
• pnpm exec oxfmt --check apps/web/src/lib/ai-gateway/providers/vercel/mapModelIdToVercel.ts apps/web/src/lib/ai-gateway/providers/vercel/mapModelIdToVercel.test.ts
• git diff --check

[kilo-code-bot]

Code Review Summary

Status: No Issues Found | Recommendation: Merge

Executive Summary

Reviewed the Vercel provider-option prep for Anthropic/OpenAI model variants; the refactor is logically sound, the mapModelIdToVercel edits net to a clean no-op across commits, and behavior is covered by the added tests, so no high-confidence issues were found in the changed code.

Files Reviewed (5 files)

• apps/web/src/lib/ai-gateway/providers/anthropic.constants.ts
• apps/web/src/lib/ai-gateway/providers/openai.ts
• apps/web/src/lib/ai-gateway/providers/openrouter/types.ts
• apps/web/src/lib/ai-gateway/providers/vercel/index.test.ts
• apps/web/src/lib/ai-gateway/providers/vercel/index.ts

---

_{Reviewed by claude-sonnet-5 · Input: 36 · Output: 11.2K · Cached: 883.4K}

_{Review guidance: REVIEW.md from base branch main}