ci: require complete 42585-row staging and zero source duplicates

.github/workflows/everest-type-first.yml

@@ -40,7 +40,7 @@ jobs:
             --input data/nomenclature.csv \
             --output output/everest_nomenclature_type_first.sqlite \
             --report output/type_first_report.json \
-            --min-rows 40000 --min-brands 74
+            --min-rows 42585 --min-brands 74
       - name: Check complete and non-promoted data
         run: |
           python - <<'PY'
@@ -48,7 +48,8 @@ jobs:
           data = json.load(open('output/type_first_report.json'))
           c = sqlite3.connect('output/everest_nomenclature_type_first.sqlite')
           assert data['brands'] >= 74
-          assert data['rows'] >= 40000
+          assert data['rows'] >= 42585
+          assert data['duplicate_source_rows'] == 0
           assert c.execute('PRAGMA integrity_check').fetchone()[0] == 'ok'
           assert c.execute('SELECT COUNT(*) FROM stage_designations').fetchone()[0] == data['rows']
           assert c.execute("SELECT COUNT(*) FROM stage_designations WHERE semantic_status != 'UNVERIFIED'").fetchone()[0] == 0