Copilot doesn't use model reasoning definitions and defaults when used via OpenRouter

Does this issue occur when all extensions are disabled?: Yes/No

- VS Code Version: 1.137.0
- OS Version: Both Windows / macOS

Steps to Reproduce:

1. Open "Manage Language " - "Add Models", select "OpenRouter"
2. Pick models from OpenRouter Group
3. The "Thinking Effort" dropdown in the model picker uses a hard-coded fixed list (low/medium/high) and does not respect the default thinking intensity that OpenRouter provides per model.

OpenRouter's `GET /api/v1/models` endpoint ( https://openrouter.ai/api/v1/models ) returns a `reasoning` object per
model, e.g.:

```json
{
"id": "openai/gpt-5.6-luna-pro",
"reasoning": {
"supported_efforts": [
"max",
"xhigh",
"high",
"medium",
"low",
"none"
],
"default_effort": "medium"
}
}
```

I'd like to work on this and submit a PR. :)