fix: correct Gemini 2.5 Flash pricing (#4836)

[daniel-lxs]

Description

Fixes #4836

This PR corrects the pricing for the Gemini 2.5 Flash model to match the official Google AI pricing documentation.

Changes Made

• Updated gemini-2.5-flash pricing in packages/types/src/providers/gemini.ts:

  • Input price: $0.15/M → $0.30/M tokens
  • Output price: $0.60/M → $2.50/M tokens
  • Cache reads price: $0.0375/M → $0.075/M tokens (matching text/image/video caching)
  • Cache writes price: kept at $1.0/M tokens (storage price)

• Updated gemini-2.5-flash pricing in packages/types/src/providers/vertex.ts with the same values

Testing

• All existing tests pass
• Linting passes
• Type checking passes
• Changes are minimal and focused only on pricing values

Verification of Acceptance Criteria

• Gemini 2.5 Flash model now shows correct pricing:
  • Input tokens: $0.30/M (was $0.15/M)
  • Output tokens: $2.50/M (was $0.60/M)
• Pricing matches the official source: https://ai.google.dev/gemini-api/docs/pricing#gemini-2.5-flash

Checklist

• Code follows project style guidelines
• Self-review completed
• No breaking changes
• Only pricing values were updated, no functional changes

---

Important

Update Gemini 2.5 Flash pricing in gemini.ts and vertex.ts to match official documentation.

• Pricing Updates:
  • Updated gemini-2.5-flash pricing in gemini.ts and vertex.ts:
    • Input price: $0.15/M → $0.30/M tokens
    • Output price: $0.60/M → $2.50/M tokens
    • Cache reads price: $0.0375/M → $0.075/M tokens
    • Cache writes price remains $1.0/M tokens

^{This description was created by for af99164. You can customize this summary. It will automatically update as commits are pushed.}