feat(provider): add ModelScope as OpenAI-compatible provider

[dataCenter430]

📝 Description

Add ModelScope (魔搭社区) as a new OpenAI-compatible LLM provider.

ModelScope is Alibaba's open-source AI model community and inference platform. It provides an OpenAI-compatible API at https://api-inference.modelscope.cn/v1 with 2,000 free API requests per day, making it an attractive free-tier option. Available models include Qwen3, GLM-4, DeepSeek-V3, and many other open-source models.

Changes:

• Register modelscope as an OpenAI-compatible protocol in the provider factory with default API base https://api-inference.modelscope.cn/v1
• Add ModelScope field to ProvidersConfig struct and IsEmpty() check
• Add default model entry (Qwen/Qwen3-235B-A22B-Instruct-2507) in DefaultConfig()
• Add legacy provider migration rule for backward compatibility
• Add provider creation, default API base, and migration tests
• Add ModelScope to example config (model_list + deprecated providers section)
• Add MODELSCOPE_API_KEY to .env.example
• Add ModelScope to provider tables and free-tier comparison sections in all 6 README translations

🗣️ Type of Change

• 🐞 Bug fix (non-breaking change which fixes an issue)
• ✨ New feature (non-breaking change which adds functionality)
• 📖 Documentation update
• ⚡ Code refactoring (no functional changes, no api changes)

🤖 AI Code Generation

• 🤖 Fully AI-generated (100% AI, 0% Human)
• 🛠️ Mostly AI-generated (AI draft, Human verified/modified)
• 👨‍💻 Mostly Human-written (Human lead, AI assisted or none)

🔗 Related Issue

Closes #1438

📚 Technical Context (Skip for Docs)

• Reference URL: https://modelscope.cn/docs/model-service/API-Inference/intro
• Reasoning: ModelScope provides a standard OpenAI-compatible /v1/chat/completions endpoint, so no new provider implementation is needed — it reuses the existing HTTPProvider (same as deepseek, moonshot, longcat, etc.). The model identifier format uses multiple slashes (e.g., modelscope/Qwen/Qwen3-235B-A22B-Instruct-2507), which is correctly handled by ExtractProtocol() since strings.Cut splits on the first / only (same behavior as nvidia and avian models).

🧪 Test Environment

• OS: Ubuntu
• Model/Provider: ModelScope / Qwen/Qwen3-235B-A22B-Instruct-2507
• Channels: N/A (provider-level change)

📸 Evidence (Optional)

Click to view Logs/Screenshots

Configuration example:

{
  "model_list": [
    {
      "model_name": "modelscope-qwen",
      "model": "modelscope/Qwen/Qwen3-235B-A22B-Instruct-2507",
      "api_key": "your-modelscope-access-token",
      "api_base": "https://api-inference.modelscope.cn/v1"
    }
  ]
}

Users can obtain their AccessToken at https://modelscope.cn/my/tokens

☑️ Checklist

• My code/docs follow the style of this project.
• I have performed a self-review of my own changes.
• I have updated the documentation accordingly.

[dataCenter430]

Hi, @alexhoshina
I'm so happy to contribute here as a contributor.
It's so important for me to contribute here continuously.
Could you pls review the PR when you have a chance? 🙏
Best regards

[alexhoshina]

LGTM! Thank you for your contribution!

[Orgmar]

@dataCenter430 Nice work adding ModelScope as a provider! The 2,000 free requests per day is a solid free-tier option, and reusing the existing HTTPProvider for the OpenAI-compatible endpoint keeps things clean. Good attention to detail with all the README translations too.

We have a PicoClaw Dev Group on Discord where contributors can connect and collaborate. If you're interested, send an email to [email protected] with the subject [Join PicoClaw Dev Group] + your GitHub account and we'll get you the invite link!

[dataCenter430]

LGTM! Thank you for your contribution!

Hi, @alexhoshina thanks for your review. 🙏
If you want, I'd love to contribute to picolaw continuously.
May I pick other issues up?
Thank you again

[alexhoshina]

Hi, @alexhoshina thanks for your review. 🙏 If you want, I'd love to contribute to picolaw continuously. May I pick other issues up? Thank you again

Of course! Thank you so much for your enthusiasm ♥️