Capture CLI has no fabric_item_id / workspace / datasetId / fabric_type fields #41

Summary

A Fabric + Power BI walk needs identity fields that are not in the capture schema/CLI:

• fabric_item_id (workspace item GUID)
• fabric_workspace / fabric_workspace_id
• pbi_dataset_id (report→model bind)
• datasources_status (200 vs 403)
• fabric_type (Report vs SemanticModel vs Lakehouse) vs DEKC noun (Dashboard vs Report)

We post-patched YAML after dekc_capture.py. Coverage-skeptic then flagged DEKC Dashboard nodes whose Fabric type is Report (Driver Scorecard vs Service Failure Scorecard inconsistency).

SemanticModel --tables always concept_ref(..., "tables"), so you cannot models a View or a DataLake.

Expected

python3 scripts/dekc_capture.py semantic --name "GWII - Exec Dashboard" \
  --fabric-id bbcc0c50-... --workspace data_central_ws \
  --verified false
python3 scripts/dekc_capture.py report --name "Driver Scorecard" \
  --tool powerbi --dataset-id bbcc0c50-... --fabric-type Report

Skills: Fabric Report is not automatically a DEKC Dashboard. Default SQL-endpoint models (data_central_lh) are not curated gold.

Suggested fix

Add optional identity keys to SemanticModel, Report, Dashboard, DataLake, IngestionJob schemas and CLIs. Do not require them for filesystem walks.

Found on: DEKC 0.4.1 (GitHub Latest 2026-08-24), mixed PKC+SAC+DEKC second brain, Azure Fabric reverse-engineering walk.
Not a sample-knowledge issue. Repro used live capture scripts from spillwave-second-brain cache data-engineering-knowledge-capture/0.4.1/.