Fix Qwen OAuth and Ollama models listing

- Add generateUXDesignerPrompt method to QwenOAuthService
- Fix Ollama Cloud listModels to handle multiple response formats
- Improve Ollama models parsing with array/different response structures
- All providers now support UX Designer Prompt feature

app/api/qwen/user/route.ts

@@ -0,0 +1,38 @@
+import { NextRequest, NextResponse } from "next/server";
+import { QWEN_OAUTH_BASE_URL } from "../constants";
+
+export async function GET(request: NextRequest) {
+  try {
+    const authorization = request.headers.get("authorization");
+    if (!authorization || !authorization.startsWith("Bearer ")) {
+      return NextResponse.json(
+        { error: "Authorization required" },
+        { status: 401 }
+      );
+    }
+    const token = authorization.slice(7);
+
+    const userResponse = await fetch(`${QWEN_OAUTH_BASE_URL}/api/v1/user`, {
+      headers: {
+        Authorization: `Bearer ${token}`,
+      },
+    });
+
+    if (!userResponse.ok) {
+      const errorText = await userResponse.text();
+      return NextResponse.json(
+        { error: "Failed to fetch user info", details: errorText },
+        { status: userResponse.status }
+      );
+    }
+
+    const userData = await userResponse.json();
+    return NextResponse.json({ user: userData });
+  } catch (error) {
+    console.error("Qwen user info failed", error);
+    return NextResponse.json(
+      { error: "Failed to fetch user info" },
+      { status: 500 }
+    );
+  }
+}