feat(server): 票据文件夹资产缓存与文档预览统一生成

- 新增 document_preview 模块，DocumentPreviewAssets 统一处理 data URL 解码、pdftoppm PNG 预览生成（poppler-data 编码）、renderer_id 标识
- receipt_folder 服务复用预览生成，缓存票据资产并提供清理；删除票据时保留已关联报销单的附件副本
- document_intelligence 新增票据预览/资产缓存接入与字段提取增强；ocr 抽取复用预览工具，附件分析/文档/操作/展示四个子模块同步适配
- receipt_folder 端点补充资产缓存头，补/扩 document_intelligence、ocr_endpoints、ocr_service、receipt_folder_service、reimbursement_endpoints 测试，新增 attachment_analysis 回归测试

server/tests/test_reimbursement_endpoints.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import base64
+import json
 from collections.abc import Generator
 from datetime import UTC, date, datetime
 from decimal import Decimal
@@ -19,6 +20,7 @@ from app.models.organization import OrganizationUnit
 from app.models.risk_observation import RiskObservation, RiskObservationFeedback
 from app.models.role import Role
 from app.schemas.ocr import OcrRecognizeBatchRead, OcrRecognizeDocumentRead
+from app.services.document_preview import DocumentPreviewAssets
 from app.services.expense_claim_attachment_storage import ExpenseClaimAttachmentStorage
 from app.services.ocr import OcrService
 
@@ -686,6 +688,9 @@ def test_claim_item_pdf_attachment_preview_returns_generated_image(monkeypatch,
     meta_payload = upload_response.json()["attachment"]
     assert meta_payload["preview_kind"] == "image"
     assert meta_payload["preview_url"].endswith(f"/reimbursements/claims/{claim_id}/items/{item_id}/attachment/preview")
+    meta_path = next(tmp_path.rglob("invoice.pdf.meta.json"))
+    stored_meta = json.loads(meta_path.read_text(encoding="utf-8"))
+    assert stored_meta["preview_rendered_with"] == DocumentPreviewAssets.PDF_RENDERER_ID
 
     preview_response = client.get(
         f"/api/v1/reimbursements/claims/{claim_id}/items/{item_id}/attachment/preview",