feat: Implement automated keyword-based news pipeline scheduler

- Add multi-threaded keyword scheduler for periodic news collection
- Create Keyword Manager API for CRUD operations and monitoring
- Implement automatic pipeline triggering (RSS → Google → AI → Translation)
- Add thread status monitoring and dynamic keyword management
- Support priority-based execution and configurable intervals
- Add comprehensive scheduler documentation guide
- Default keywords: AI, 테크놀로지, 경제, 블록체인

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

services/pipeline/test_pipeline.py

@@ -0,0 +1,118 @@
+#!/usr/bin/env python3
+"""
+Pipeline Test Script
+파이프라인 전체 플로우를 테스트하는 스크립트
+"""
+import asyncio
+import json
+from datetime import datetime
+from motor.motor_asyncio import AsyncIOMotorClient
+import redis.asyncio as redis
+from shared.models import KeywordSubscription, PipelineJob
+
+async def test_pipeline():
+    """파이프라인 테스트"""
+    
+    # MongoDB 연결
+    mongo_client = AsyncIOMotorClient("mongodb://mongodb:27017")
+    db = mongo_client.pipeline
+    
+    # Redis 연결
+    redis_client = redis.Redis(host='redis', port=6379, decode_responses=True)
+    
+    # 1. 테스트 키워드 추가
+    test_keyword = KeywordSubscription(
+        keyword="전기차",
+        language="ko",
+        schedule="*/1 * * * *",  # 1분마다 (테스트용)
+        is_active=True,
+        is_priority=True,
+        rss_feeds=[
+            "https://news.google.com/rss/search?q=전기차&hl=ko&gl=KR&ceid=KR:ko",
+            "https://news.google.com/rss/search?q=electric+vehicle&hl=en&gl=US&ceid=US:en"
+        ],
+        categories=["technology", "automotive", "environment"],
+        owner="test_user"
+    )
+    
+    # MongoDB에 저장
+    await db.keyword_subscriptions.replace_one(
+        {"keyword": test_keyword.keyword},
+        test_keyword.dict(),
+        upsert=True
+    )
+    print(f"✅ 키워드 '{test_keyword.keyword}' 추가 완료")
+    
+    # 2. 즉시 파이프라인 트리거 (스케줄러를 거치지 않고 직접)
+    job = PipelineJob(
+        keyword_id=test_keyword.keyword_id,
+        keyword=test_keyword.keyword,
+        stage="rss_collection",
+        data={
+            "rss_feeds": test_keyword.rss_feeds,
+            "categories": test_keyword.categories
+        },
+        priority=1 if test_keyword.is_priority else 0
+    )
+    
+    # Redis 큐에 직접 추가 (QueueMessage 형식으로)
+    from shared.queue_manager import QueueMessage
+    message = QueueMessage(
+        queue_name="rss_collection",
+        job=job
+    )
+    await redis_client.lpush("queue:rss_collection", message.json())
+    print(f"✅ 작업을 RSS Collection 큐에 추가: {job.job_id}")
+    
+    # 3. 파이프라인 상태 모니터링
+    print("\n📊 파이프라인 실행 모니터링 중...")
+    print("각 단계별 로그를 확인하려면 다음 명령을 실행하세요:")
+    print("  docker-compose logs -f pipeline-rss-collector")
+    print("  docker-compose logs -f pipeline-google-search")
+    print("  docker-compose logs -f pipeline-ai-summarizer")
+    print("  docker-compose logs -f pipeline-translator")
+    print("  docker-compose logs -f pipeline-image-generator")
+    print("  docker-compose logs -f pipeline-article-assembly")
+    
+    # 큐 상태 확인
+    for i in range(10):
+        await asyncio.sleep(5)
+        
+        # 각 큐의 길이 확인
+        queues = [
+            "queue:rss_collection",
+            "queue:google_search", 
+            "queue:ai_summarization",
+            "queue:translation",
+            "queue:image_generation",
+            "queue:article_assembly"
+        ]
+        
+        print(f"\n[{datetime.now().strftime('%H:%M:%S')}] 큐 상태:")
+        for queue in queues:
+            length = await redis_client.llen(queue)
+            if length > 0:
+                print(f"  {queue}: {length} 작업 대기 중")
+    
+    # 4. 최종 결과 확인
+    print("\n📄 MongoDB에서 생성된 기사 확인 중...")
+    articles = await db.articles.find({"keyword": test_keyword.keyword}).to_list(length=5)
+    
+    if articles:
+        print(f"✅ {len(articles)}개의 기사 생성 완료!")
+        for article in articles:
+            print(f"\n제목: {article.get('title', 'N/A')}")
+            print(f"ID: {article.get('article_id', 'N/A')}")
+            print(f"생성 시간: {article.get('created_at', 'N/A')}")
+            print(f"처리 시간: {article.get('processing_time', 'N/A')}초")
+            print(f"이미지 수: {len(article.get('images', []))}")
+    else:
+        print("⚠️ 아직 기사가 생성되지 않았습니다. 조금 더 기다려주세요.")
+    
+    # 연결 종료
+    await redis_client.close()
+    mongo_client.close()
+
+if __name__ == "__main__":
+    print("🚀 파이프라인 테스트 시작")
+    asyncio.run(test_pipeline())