feat(worker): summarize sliding window — 50k chunk + cumulative carry-over

P3 of family-adaptive-bengio (Mac mini 4-lever bundle).

50k 초과 input 은 CHUNK_SIZE=50000 단위로 N 분할 + cumulative carry-over (prev chunk summary 를 다음 chunk prompt 에 prefix). 50k 이하 input = 기존 동작 (변동 0). 첫 chunk = client.summarize() legacy / 후속 chunk = call_primary + SUMMARY_PROMPT_CONTINUATION. log trace: single vs sliding chunk N/M done.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

app/workers/summarize_worker.py

@@ -1,4 +1,8 @@
-"""요약 전용 워커 — 뉴스 등 classify 불필요한 문서의 AI 요약만 생성"""
+"""요약 전용 워커 — 뉴스 등 classify 불필요한 문서의 AI 요약만 생성.
+
+P3 of family-adaptive-bengio (2026-05-23): 50k 초과 input 은 sliding window
+(cumulative carry-over) 로 분할 처리. 50k 이하 input 은 기존 동작 유지.
+"""
 
 from datetime import datetime, timezone
 
@@ -10,6 +14,12 @@ from models.document import Document
 
 logger = setup_logger("summarize_worker")
 
+CHUNK_SIZE = 50000
+SUMMARY_PROMPT_CONTINUATION = (
+    "이전 부분 요약:\n{prior}\n\n다음 부분:\n{text}\n\n"
+    "위 두 정보를 합쳐 전체 문서를 500자 이내로 요약해주세요."
+)
+
 
 async def process(document_id: int, session: AsyncSession) -> None:
     """문서 AI 요약 생성 (분류 없이 요약만)"""
@@ -26,10 +36,38 @@ async def process(document_id: int, session: AsyncSession) -> None:
 
     client = AIClient()
     try:
-        summary = await client.summarize(doc.extracted_text[:50000])
+        text = doc.extracted_text
+        total_chars = len(text)
+        if total_chars <= CHUNK_SIZE:
+            summary = await client.summarize(text)
+            logger.info(
+                f"[요약] document_id={document_id}: single chunk ({total_chars}자)"
+            )
+        else:
+            chunks = [text[i:i + CHUNK_SIZE] for i in range(0, total_chars, CHUNK_SIZE)]
+            logger.info(
+                f"[요약] document_id={document_id}: sliding window {len(chunks)} chunk "
+                f"(total {total_chars}자, chunk_size={CHUNK_SIZE})"
+            )
+            carry = ""
+            for idx, chunk in enumerate(chunks):
+                if idx == 0:
+                    partial = await client.summarize(chunk)
+                else:
+                    prompt = SUMMARY_PROMPT_CONTINUATION.format(prior=carry, text=chunk)
+                    partial = await client.call_primary(prompt)
+                carry = strip_thinking(partial)
+                logger.info(
+                    f"[요약] document_id={document_id}: chunk {idx + 1}/{len(chunks)} done "
+                    f"(in={len(chunk)}자, carry={len(carry)}자)"
+                )
+            summary = carry
+
         doc.ai_summary = strip_thinking(summary)
         doc.ai_model_version = "qwen3.5-35b-a3b"
         doc.ai_processed_at = datetime.now(timezone.utc)
-        logger.info(f"[요약] document_id={document_id}: {len(doc.ai_summary)}자")
+        logger.info(
+            f"[요약] document_id={document_id}: {len(doc.ai_summary)}자 final"
+        )
     finally:
         await client.close()