Add CoderPlan provider

[onepaperbox]

Summary

Adds CoderPlan as a provider — an OpenAI-compatible LLM API relay for Chinese developers.

• 32 models: Claude (Opus 4.6–4.8, Sonnet 4.6, Haiku 4.5), GPT (5.5, 5.4, 5.4-mini), Gemini (2.5 Flash/Pro, 3 Flash/Pro, 3.1 Flash/Pro), DeepSeek (V4 Flash/Pro), GLM (5, 5.1), Kimi (K2.5, K2.6), Qwen (3.6-plus, 3.7-max, 3.7-plus), MiniMax (M2.5, M2.7, M3 + highspeed), MiMo (V2.5, V2.5-Pro), HY3
• Pricing: pay-per-use at ~0.7x official rates, ¥10 minimum top-up, Alipay/WeChat
• API endpoint: https://api.coderplan.ai/v1 (OpenAI-compatible)
• Docs: https://coderplan.ai/docs/setup

All model definitions use base_model to inherit capabilities from the canonical base providers, with CoderPlan-specific pricing overrides.

Checklist

• provider.toml follows existing schema
• Model files use base_model from canonical base providers
• Pricing data matches published rates at https://coderplan.ai/pricing
• Logo SVG uses currentColor with no fixed dimensions

[onepaperbox]

Hi! I noticed the Validate Models workflow requires approval to run for first-time contributors. Could a maintainer please approve the CI run? Thanks!

[rekram1-node]

AUTOMATED REVIEW: This PR is not yet compliant with the model contribution requirements.

32 of the 33 added CoderPlan models inherit reasoning = true through base_model but omit provider-specific reasoning_options. Please add explicit reasoning_options to every added model except providers/coderplan/models/gpt-5-mini.toml (use an empty array where the provider exposes no reasoning control).

The base_model usage and new provider logo pass the requested requirements.