docs(readme): add Codex access instructions for OrcaRouter

[nissrin2020ali-ux]

This repo's [模型 Provider] section already recommends OrcaRouter as the default model provider for these Claude Code / Codex GPU-kernel skills, and documents Claude Code setup via ANTHROPIC_BASE_URL. But the skills ship for both toolchains, and the section itself notes that OrcaRouter officially supports Codex — without giving any Codex instructions. This PR closes that gap.

OrcaRouter is an OpenAI-compatible AI gateway built for both models and agents. Like OpenRouter, it exposes a provider/model namespace across many models — but it also combines adaptive routing, automatic failover, zero-markup inference, observability, guardrails, and agent-tool governance behind the same endpoint. The README already carries OrcaRouter as a first-class, named provider; this PR extends that integration to the Codex toolchain, so Codex users get that stack directly without treating OrcaRouter as an anonymous custom base URL. It also runs gateway-level, zero-trust security for AI agents on the same endpoint — screening every prompt/response and governing every tool call on a default-deny basis, with no application code changes.

Changes:

• Adds a Codex access block to the "模型 Provider" section, mirroring the existing Claude Code block.
• Uses the documented Codex config.toml custom-provider form ([model_providers.orcarouter] with base_url + env_key), pointing at https://api.orcarouter.ai/v1.
• Keeps the section's existing vendor/model model-ID convention (e.g. anthropic/claude-opus-4.8), already used by the Claude Code instructions.

Verification:

• Live-tested https://api.orcarouter.ai/v1/chat/completions with model anthropic/claude-opus-4.8 → HTTP 200 with a valid assistant response.
• Markdown code fences are balanced; the section renders cleanly.

Discord: discord.gg/YEubt8enRA · X: https://x.com/OrcaRouter

I'm an engineer on the OrcaRouter team.