Add OpenAI GPT-5.3 chat latest and GPT-5.4 model support

[PeterDaveHello]

Description

This PR adds OpenAI Native support for gpt-5.3-chat-latest and gpt-5.4.

Implementation details:

• Added gpt-5.3-chat-latest to openAiNativeModels with chat model limits and pricing:
  • maxTokens: 16_384
  • contextWindow: 128_000
  • inputPrice: 1.75
  • outputPrice: 14.0
  • cacheReadsPrice: 0.175
  • supportsImages: true
  • supportsPromptCache: true
• Kept gpt-5.3-chat-latest aligned with existing chat-latest conventions by not explicitly setting supportsTemperature.
• Added gpt-5.4 to openAiNativeModels with current limits, pricing, and capabilities:
  • maxTokens: 128_000
  • contextWindow: 1_050_000
  • inputPrice: 2.5
  • cacheReadsPrice: 0.25
  • outputPrice: 15.0
  • supportsImages: true
  • supportsPromptCache: true
  • supportsVerbosity: true
  • supportsReasoningEffort: ["none", "low", "medium", "high", "xhigh"]
  • reasoningEffort: "none"
• Added GPT-5.4 long-context cost handling for > 272_000 input tokens:
  • applies only to default and flex service tiers
  • applies 2x uncached input pricing and 1.5x output pricing
  • does not apply the surcharge to priority or cached input reads
• Added coverage in openai-native tests for:
  • model resolution via getModel()
  • Responses API request bodies for gpt-5.3-chat-latest and gpt-5.4
  • GPT-5.4 default reasoning payload (reasoning.effort = "none")
  • prompt cache retention behavior
  • GPT-5.4 usage-cost normalization across base, long-context, and priority-tier scenarios
• Added changesets for patch releases:
  • .changeset/gpt-5-3-chat-latest.md
  • .changeset/gpt-5-4.md

Files changed:

• packages/types/src/model.ts
• packages/types/src/providers/openai.ts
• src/shared/cost.ts
• src/api/providers/openai-native.ts
• src/api/providers/__tests__/openai-native.spec.ts
• src/api/providers/__tests__/openai-native-usage.spec.ts
• src/utils/__tests__/cost.spec.ts
• .changeset/gpt-5-3-chat-latest.md
• .changeset/gpt-5-4.md

Test Procedure

Run targeted tests for the affected provider and cost specs:

pnpm --filter roo-cline exec vitest run api/providers/__tests__/openai-native.spec.ts api/providers/__tests__/openai-native-usage.spec.ts utils/__tests__/cost.spec.ts

Observed result:

• Test Files 3 passed (3)
• Tests 89 passed (89)

Notes:

• Running pnpm --filter roo-cline test -- ... triggers pretest and a broader validation surface, so the targeted vitest run command above was used for focused verification.

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

N/A (no UI changes)

Documentation Updates

• No documentation updates are required.
• Yes, documentation updates are required.