fix(openrouter): pass GPT-5 reasoning effort and include_reasoning to OpenRouter

[roomote]

This PR ensures the Model Reasoning Effort is correctly passed to OpenRouter for GPT‑5 models and that reasoning content is requested in the stream.

Changes:

• Set include_reasoning=true for GPT‑5 requests in createMessage and completePrompt (OpenRouterHandler.createMessage(), OpenRouterHandler.completePrompt()).
• Default GPT‑5 reasoning effort to "medium" when enabled but unspecified, while honoring explicit settings (including "minimal") in OpenRouterHandler.getModel().
• Added tests asserting:
  • include_reasoning true and reasoning: { effort: "minimal" } are passed for GPT‑5 via OpenRouter (spec).
  • default reasoning effort "medium" is applied when enabled but not specified (spec).

Context:

• OpenRouter handler previously relied on upstream reasoning resolution. This PR adds GPT‑5–specific handling to guarantee effort is present and reasoning content is included for streaming.

Validation:

• Unit tests pass for updated OpenRouter behavior.
• Lint and type checks pass via turborepo pipeline.

Security:

• No secrets logged. Error handling remains contained.

---

Important

Enhance OpenRouter integration for GPT-5 by ensuring reasoning efforts are passed and included, with default effort set when unspecified.

• Behavior:
  • Set include_reasoning=true for GPT-5 requests in OpenRouterHandler.createMessage() and OpenRouterHandler.completePrompt().
  • Default GPT-5 reasoning effort to "medium" when enabled but unspecified in OpenRouterHandler.getModel().
• Tests:
  • Added tests in openrouter.spec.ts to assert include_reasoning and reasoning effort for GPT-5.
  • Tests for default reasoning effort "medium" when not specified.
• Context:
  • Previously relied on upstream reasoning resolution; now ensures effort is present and reasoning content is included for streaming.

^{This description was created by for 039e59f. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

I reviewed my own code and found it suspiciously adequate.

[roomote]

Is this logic potentially redundant? I notice we're adding GPT-5-specific reasoning handling here, but getModelParams() already calls getOpenRouterReasoning() which should handle reasoning parameters. Could we rely solely on the existing getModelParams() flow instead of duplicating the logic?

[roomote]

Consider extracting this GPT-5 check to a helper function like isGpt5Model(modelId) since the same check appears in both createMessage() (line 125) and completePrompt() (line 233). This would make future updates easier and more maintainable.

[roomote]

Good test coverage for createMessage()! Should we add a similar test for completePrompt() to verify that include_reasoning: true is also passed for GPT-5 models in that method?