feat: add Qwen3 Next 80B A3B models to chutes provider

[daniel-lxs]

Summary

This PR adds two new Qwen3 Next models to the chutes provider:

Models Added

1. Qwen3-Next-80B-A3B-Instruct

   • Fast, stable instruction-tuned model optimized for complex tasks, RAG, and tool use without thinking traces
   • Context window: 131,072 tokens
   • Max tokens: 32,768

2. Qwen3-Next-80B-A3B-Thinking

   • Reasoning-first model with structured thinking traces for multi-step problems, math proofs, and code synthesis
   • Context window: 131,072 tokens
   • Max tokens: 32,768

Pricing

Both models are set to /bin/sh input/output pricing, following the existing pattern where most chutes models (28 out of 31) have zero pricing.

Testing

• ✅ All chutes provider tests pass
• ✅ TypeScript compilation successful
• ✅ Pre-commit hooks pass

Changes

• Added model IDs to ChutesModelId type
• Added model configurations to chutesModels object with appropriate descriptions

---

Important

Add Qwen3 Next 80B A3B models to chutes provider with specific configurations and zero pricing.

• Models Added:
  • Qwen/Qwen3-Next-80B-A3B-Instruct and Qwen/Qwen3-Next-80B-A3B-Thinking added to ChutesModelId in chutes.ts.
  • Configurations for both models added to chutesModels with context window of 131,072 tokens and max tokens of 32,768.
  • Both models have zero input/output pricing.
• Testing:
  • All chutes provider tests pass.
  • TypeScript compilation successful.
  • Pre-commit hooks pass.

^{This description was created by for 825a0be. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Thank you for your contribution! I've reviewed the changes and the implementation looks clean and follows the existing patterns well. I have a few suggestions for improvement.

[roomote]

Given that this is a "Thinking" model with structured thinking traces (as mentioned in the description), should we consider adding reasoning support properties like supportsReasoningBudget: true or supportsReasoningEffort: true?

I noticed other thinking models in the codebase like "Qwen3-235B-A22B-Thinking-2507" (line 304) have similar descriptions. Would this model benefit from the same reasoning capabilities?