Docs: add AnyLLM as a documented OpenAI-compatible provider (24+ models, one key)

Summary aichat already supports the major Chinese providers individually — Deepseek, Qianwen, Moonshot, ZhipuAI, MiniMax, Ernie. But for users outside China, each one requires a separate account with a Chinese phone number and Alipay, which is a real adoption blocker.

I built AnyLLM, an OpenAI-compatible API gateway that exposes 24+ Chinese and international models behind a single key: DeepSeek V4, Qwen3.7-Plus, GLM-5.2, Kimi K2.5, MiniMax M3 — alongside GPT-5 and Gemini. International cards and crypto accepted, pass-through pricing with zero markup, built-in routing and failover.

Why this fits aichat aichat already handles "any OpenAI-Compatible API provider" — AnyLLM works today via a custom client entry, no new code needed. One API key replaces five separate Chinese provider accounts. The LLM Arena and Playground get more interesting when 24+ models are reachable through one endpoint — side-by-side comparison across Chinese and Western frontier models. Proposal Either of these, whichever fits the project best:

A short wiki section: "Using an OpenAI-compatible gateway (AnyLLM)" with a sample config. A preset entry in models.yaml so users can pick AnyLLM models out of the box. Happy to open a PR for either — docs page or models.yaml entry.

(Context: I'm the developer of AnyLLM — happy to provide a sandbox key for testing.)