feat(channel-monitor): 内置 OpenAI 检测模板

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>

backend/migrations/139_seed_openai_monitor_templates.sql

@@ -0,0 +1,47 @@
+-- Migration: 138_seed_openai_monitor_templates
+-- 内置 OpenAI 渠道监控模板。重点是把协议模式显式化：
+--   1) OpenAI-compatible 使用 Chat Completions payload
+--   2) Responses / 本站自检 使用 Responses payload，默认 body 由后端 adapter 填入 instructions + input
+-- 所有模板都可直接选择；ON CONFLICT 保证重复部署不覆盖用户编辑。
+
+INSERT INTO channel_monitor_request_templates (
+    name, provider, api_mode, description, extra_headers, body_override_mode, body_override
+)
+VALUES
+(
+    'OpenAI Compatible 默认检测',
+    'openai',
+    'chat_completions',
+    '适用于大多数 OpenAI-compatible 上游：POST /v1/chat/completions，后端自动生成 messages 数学 challenge。',
+    '{}'::jsonb,
+    'off',
+    NULL
+),
+(
+    'OpenAI Compatible 低 token 检测',
+    'openai',
+    'chat_completions',
+    '仍走 /v1/chat/completions，仅把 max_tokens 调低；model/messages/stream 由后端保护，避免误伤 challenge。',
+    '{}'::jsonb,
+    'merge',
+    '{"max_tokens": 20}'::jsonb
+),
+(
+    'OpenAI Responses / 本站自检',
+    'openai',
+    'responses',
+    '适用于本站或原生 Responses API：POST /v1/responses，默认 payload 自动带 instructions 与 input，避免 Instructions are required。',
+    '{}'::jsonb,
+    'off',
+    NULL
+),
+(
+    'OpenAI Responses 低 token 检测',
+    'openai',
+    'responses',
+    '仍走 /v1/responses，仅把 max_output_tokens 调低；instructions/input/model/stream 由后端保护。',
+    '{}'::jsonb,
+    'merge',
+    '{"max_output_tokens": 20}'::jsonb
+)
+ON CONFLICT (provider, name) DO NOTHING;