fix: Change base_url to full_url

extensions/inference-openai-extension/src/@types/global.d.ts

@@ -3,7 +3,7 @@ import { Model } from "@janhq/core";
 declare const MODULE: string;
 
 declare interface EngineSettings {
-  base_url?: string;
+  full_url?: string;
   api_key?: string;
 }
 

extensions/inference-openai-extension/src/helpers/sse.ts

@@ -22,7 +22,7 @@ export function requestInference(
       // Model parameters spreading
       ...model.parameters,
     });
-    fetch(`${engine.base_url}/chat/completions`, {
+    fetch(`${engine.full_url}`, {
       method: "POST",
       headers: {
         "Content-Type": "application/json",

extensions/inference-openai-extension/src/index.ts

@@ -37,7 +37,7 @@ export default class JanInferenceOpenAIExtension implements InferenceExtension {
   private static _currentModel: OpenAIModel;
 
   private static _engineSettings: EngineSettings = {
-    base_url: "https://api.openai.com/v1",
+    full_url: "https://api.openai.com/v1/chat/completion",
     api_key: "sk-<your key here>",
   };