feat: add question insight analytics

ai_knowledge_base_v2/apps/backend/tests/test_admin_pagination.py

@@ -1,9 +1,10 @@
 from sqlalchemy import create_engine
+from datetime import datetime, timedelta
 from sqlalchemy.orm import Session
 from sqlalchemy.pool import StaticPool
 
 from app.api.admin_agent_records import attention_list, retrieval_logs
-from app.api.admin_records import ai_logs, chat_detail, chat_messages
+from app.api.admin_records import ai_logs, chat_detail, chat_messages, question_insights
 from app.api.admin_users import list_users
 from app.models import Base
 from app.models.chat import ChatMessage, ChatSession
@@ -106,3 +107,43 @@ def test_retrieval_and_attention_lists_are_paginated():
         attention = attention_list(priority="", statusValue="", page=3, pageSize=10, db=db, current_admin=object())["data"]
         assert retrieval["total"] == attention["total"] == 21
         assert len(retrieval["items"]) == len(attention["items"]) == 1
+
+
+def test_question_insights_clean_and_cluster_similar_user_questions():
+    with _database() as db:
+        user_a = User(id=1, phone="13800000000", name="学员A", daily_chat_limit=10)
+        user_b = User(id=2, phone="13800000001", name="学员B", daily_chat_limit=10)
+        session_a = ChatSession(id=1, user_id=1, title="问题", message_count=3)
+        session_b = ChatSession(id=2, user_id=2, title="问题", message_count=2)
+        now = datetime(2026, 7, 31, 10, 0, 0)
+        db.add_all([user_a, user_b, session_a, session_b])
+        db.add_all(
+            [
+                ChatMessage(id=1, session_id=1, user_id=1, role="user", content="老师你好，心光有哪些作业？", created_at=now),
+                ChatMessage(id=2, session_id=1, user_id=1, role="assistant", content="回答", created_at=now + timedelta(seconds=1)),
+                ChatMessage(id=3, session_id=2, user_id=2, role="user", content="请问心光都有什么功课", created_at=now + timedelta(minutes=1)),
+                ChatMessage(id=4, session_id=2, user_id=2, role="user", content="谢谢老师", created_at=now + timedelta(minutes=2)),
+                ChatMessage(id=5, session_id=1, user_id=1, role="user", content="1、回放在哪里看？\n2、上课链接在哪", created_at=now + timedelta(minutes=3)),
+            ]
+        )
+        db.commit()
+
+        response = question_insights(
+            dateFrom=None,
+            dateTo=None,
+            minCount=2,
+            maxMessages=100,
+            page=1,
+            pageSize=10,
+            db=db,
+            current_admin=object(),
+        )
+
+        data = response["data"]
+        assert data["summary"]["scannedMessages"] == 4
+        assert data["summary"]["filteredMessages"] == 1
+        assert data["summary"]["cleanedQuestions"] == 4
+        assert data["total"] == 1
+        assert data["items"][0]["count"] == 2
+        assert "心光" in data["items"][0]["title"]
+        assert data["items"][0]["userCount"] == 2