feat: channel gateway implementation and multi-bot refactor

- Refactor channel manager with support for Sender/Receiver interfaces and hot-swappable adapters.
- Implement identity routing and pre-authentication logic for inbound messages.
- Update database schema to support bot pre-auth keys and extended channel session metadata.
- Add Telegram and Feishu channel configuration and adapter enhancements.
- Update Swagger documentation and internal handlers for channel management.

Co-authored-by: Cursor <cursoragent@cursor.com>

agent/src/model.ts

@@ -5,17 +5,22 @@ import { createGoogleGenerativeAI } from '@ai-sdk/google'
 import { ClientType, ModelConfig } from './types'
 
 export const createModel = (model: ModelConfig) => {
-  const apiKey = model.apiKey.toLowerCase().trim()
-  const baseURL = model.baseUrl.toLowerCase().trim()
-  const modelId = model.modelId.toLowerCase().trim()
-  const clients = {
-    [ClientType.OpenAI]: createOpenAI,
-    [ClientType.OpenAICompatible]: createOpenAI,
-    [ClientType.Anthropic]: createAnthropic,
-    [ClientType.Google]: createGoogleGenerativeAI,
+  const apiKey = model.apiKey.trim()
+  const baseURL = model.baseUrl.trim()
+  const modelId = model.modelId.trim()
+
+  switch (model.clientType) {
+    case ClientType.OpenAI:
+    case ClientType.OpenAICompatible: {
+      const provider = createOpenAI({ apiKey, baseURL })
+      // Use .chat() to call /chat/completions (not /responses which only OpenAI supports)
+      return provider.chat(modelId)
+    }
+    case ClientType.Anthropic:
+      return createAnthropic({ apiKey, baseURL })(modelId)
+    case ClientType.Google:
+      return createGoogleGenerativeAI({ apiKey, baseURL })(modelId)
+    default:
+      return createAiGateway({ apiKey, baseURL })(modelId)
   }
-  return (clients[model.clientType] ?? createAiGateway)({
-    apiKey,
-    baseURL,
-  })(modelId)
 }