feat(asl/extraction): Complete Tool 3 M1+M2 - skeleton pipeline and HITL workbench

M1 Skeleton Pipeline:
- Scatter-dispatch + Aggregator polling pattern (PgBoss)
- PKB ACL bridge (PkbBridgeService -> PkbExportService DTOs)
- ExtractionSingleWorker with DeepSeek-V3 LLM extraction
- PermanentExtractionError for non-retryable failures
- Phantom Retry Guard (idempotent worker)
- 3-step minimal frontend (Setup -> Progress -> Workbench)
- 4 new DB tables (extraction_templates, project_templates, tasks, results)
- 3 system templates seed (RCT, Cohort, QC)
- M1 integration test suite

M2 HITL Workbench:
- MinerU VLM integration for high-fidelity table extraction
- XML-isolated DynamicPromptBuilder with flat JSON output template
- fuzzyQuoteMatch validator (3-tier confidence scoring)
- SSE real-time logging via ExtractionEventBus
- Schema-driven ExtractionDrawer (dynamic field rendering from template)
- Excel wide-table export with flattenModuleData normalization
- M2 integration test suite

Critical Fixes (data normalization):
- DynamicPromptBuilder: explicit flat key-value output format with example
- ExtractionExcelExporter: handle both array and flat data formats
- ExtractionDrawer: schema-driven rendering instead of hardcoded fields
- ExtractionValidator: array-format quote verification support
- SSE route: Fastify register encapsulation to bypass auth for EventSource
- LLM JSON sanitizer: strip illegal control chars before JSON.parse

Also includes: RVW stats verification spec, SSA expert config guide

Tested: M1 pipeline test + M2 HITL test + manual frontend verification
Co-authored-by: Cursor <cursoragent@cursor.com>

backend/src/modules/pkb/services/PkbExportService.ts

@@ -0,0 +1,96 @@
+import { prisma } from '../../../config/database.js';
+
+/**
+ * PKB 数据导出服务（PKB 模块维护）
+ * 
+ * ACL 防腐层出口：返回纯 DTO 对象，不暴露 Prisma 类型。
+ * 消费方：ASL PkbBridgeService
+ */
+
+export interface PkbDocumentExportDTO {
+  documentId: string;
+  storageKey: string;
+  filename: string;
+  extractedText: string | null;
+  fileSizeBytes: number;
+}
+
+export interface PkbKnowledgeBaseExportDTO {
+  id: string;
+  name: string;
+  fileCount: number;
+}
+
+class PkbExportServiceImpl {
+  async listKnowledgeBases(userId: string): Promise<PkbKnowledgeBaseExportDTO[]> {
+    const kbs = await prisma.knowledgeBase.findMany({
+      where: { userId },
+      select: {
+        id: true,
+        name: true,
+        fileCount: true,
+      },
+      orderBy: { updatedAt: 'desc' },
+    });
+
+    return kbs.map(kb => ({
+      id: kb.id,
+      name: kb.name,
+      fileCount: kb.fileCount,
+    }));
+  }
+
+  async listPdfDocuments(kbId: string): Promise<PkbDocumentExportDTO[]> {
+    const docs = await prisma.document.findMany({
+      where: {
+        kbId,
+        fileType: { in: ['pdf', 'application/pdf'] },
+      },
+      select: {
+        id: true,
+        storageKey: true,
+        filename: true,
+        extractedText: true,
+        fileSizeBytes: true,
+      },
+      orderBy: { uploadedAt: 'desc' },
+    });
+
+    return docs.map(doc => ({
+      documentId: doc.id,
+      storageKey: doc.storageKey || '',
+      filename: doc.filename,
+      extractedText: doc.extractedText,
+      fileSizeBytes: Number(doc.fileSizeBytes || 0),
+    }));
+  }
+
+  async getDocumentForExtraction(documentId: string): Promise<PkbDocumentExportDTO> {
+    const doc = await prisma.document.findUnique({
+      where: { id: documentId },
+      select: {
+        id: true,
+        storageKey: true,
+        filename: true,
+        extractedText: true,
+        fileSizeBytes: true,
+      },
+    });
+
+    if (!doc) {
+      const err = new Error(`PKB Document not found: ${documentId}`);
+      (err as any).name = 'PkbDocumentNotFoundError';
+      throw err;
+    }
+
+    return {
+      documentId: doc.id,
+      storageKey: doc.storageKey || '',
+      filename: doc.filename,
+      extractedText: doc.extractedText,
+      fileSizeBytes: Number(doc.fileSizeBytes || 0),
+    };
+  }
+}
+
+export const pkbExportService = new PkbExportServiceImpl();