Request

Please add openai-codex/gpt-5.2 to the xhigh thinking whitelist (XHIGH_MODEL_REFS).

Context

In dist/auto-reply/thinking.js, xhigh is currently restricted to a small hard-coded set (e.g. openai/gpt-5.2, openai-codex/gpt-5.2-codex, openai-codex/gpt-5.1-codex).

However, many setups (including ours) use the OpenAI Codex provider and run the default/allowed model as openai-codex/gpt-5.2 (alias gpt52). When setting thinking="xhigh", we hit:

Error: Thinking level "xhigh" is only supported for openai/gpt-5.2, openai-codex/gpt-5.2-codex or openai-codex/gpt-5.1-codex.

Proposal

Add openai-codex/gpt-5.2 to the allowlist so users on the codex provider can use xhigh with the base gpt-5.2 model (not only the -codex variant).

Why this matters

• Keeps behavior consistent across providers (openai vs openai-codex).
• Avoids confusing configuration loops where gpt52 works but xhigh is rejected.

Thanks!