feat: copilot controller (#6272)

fix CLOUD-27

packages/backend/server/src/plugins/copilot/providers/openai.ts

@@ -3,14 +3,16 @@ import assert from 'node:assert';
 import { ClientOptions, OpenAI } from 'openai';
 
 import {
-  ChatMessage,
   ChatMessageRole,
   CopilotCapability,
   CopilotProviderType,
   CopilotTextToEmbeddingProvider,
   CopilotTextToTextProvider,
+  PromptMessage,
 } from '../types';
 
+const DEFAULT_DIMENSIONS = 256;
+
 export class OpenAIProvider
   implements CopilotTextToTextProvider, CopilotTextToEmbeddingProvider
 {
@@ -50,7 +52,7 @@ export class OpenAIProvider
     return OpenAIProvider.capabilities;
   }
 
-  private chatToGPTMessage(messages: ChatMessage[]) {
+  private chatToGPTMessage(messages: PromptMessage[]) {
     // filter redundant fields
     return messages.map(message => ({
       role: message.role,
@@ -63,7 +65,7 @@ export class OpenAIProvider
     embeddings,
     model,
   }: {
-    messages?: ChatMessage[];
+    messages?: PromptMessage[];
     embeddings?: string[];
     model: string;
   }) {
@@ -106,7 +108,7 @@ export class OpenAIProvider
   // ====== text to text ======
 
   async generateText(
-    messages: ChatMessage[],
+    messages: PromptMessage[],
     model: string = 'gpt-3.5-turbo',
     options: {
       temperature?: number;
@@ -134,8 +136,8 @@ export class OpenAIProvider
   }
 
   async *generateTextStream(
-    messages: ChatMessage[],
-    model: string,
+    messages: PromptMessage[],
+    model: string = 'gpt-3.5-turbo',
     options: {
       temperature?: number;
       maxTokens?: number;
@@ -179,7 +181,7 @@ export class OpenAIProvider
       dimensions: number;
       signal?: AbortSignal;
       user?: string;
-    } = { dimensions: 256 }
+    } = { dimensions: DEFAULT_DIMENSIONS }
   ): Promise<number[][]> {
     messages = Array.isArray(messages) ? messages : [messages];
     this.checkParams({ embeddings: messages, model });
@@ -187,7 +189,7 @@ export class OpenAIProvider
     const result = await this.instance.embeddings.create({
       model: model,
       input: messages,
-      dimensions: options.dimensions,
+      dimensions: options.dimensions || DEFAULT_DIMENSIONS,
       user: options.user,
     });
     return result.data.map(e => e.embedding);