ci(validate): sparse checkout data only

Author: Seungpyo1007

.github/workflows/validate-data.yml

@@ -1,31 +1,39 @@
-name: validate-data
-
-# Self-check: every PR runs the bundled lightweight validator. Heavy checks
-# (coverage gaps, schema migrations, ingestion smoke tests) live in TechEngine.
-on:
-  pull_request:
-    paths:
-      - "data/**"
-      - "app/validate.py"
-  push:
-    branches: [main]
-    paths:
-      - "data/**"
-      - "app/validate.py"
-
-jobs:
-  self-validate:
-    runs-on: ubuntu-latest
-    steps:
+name: validate-data
+
+# Self-check: every PR runs the bundled lightweight validator. Heavy checks
+# (coverage gaps, schema migrations, ingestion smoke tests) live in TechEngine.
+on:
+  pull_request:
+    paths:
+      - "data/**"
+      - "app/validate.py"
+  push:
+    branches: [main]
+    paths:
+      - "data/**"
+      - "app/validate.py"
+
+jobs:
+  self-validate:
+    runs-on: ubuntu-latest
+    steps:
       - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
         with:
-          python-version: "3.12"
-      - name: Self-check (bundled validator)
-        run: python -m app.validate
-
-  engine-validate:
-    needs: self-validate
-    uses: GetTechAPI/TechEngine/.github/workflows/validate-data.yml@main
-    with:
-      data-ref: ${{ github.event.pull_request.head.sha || github.sha }}
+          # The committed Pages dump contains over a million generated files.
+          # Validation only needs records and the bundled validator.
+          filter: blob:none
+          sparse-checkout: |
+            app
+            data
+          sparse-checkout-cone-mode: true
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Self-check (bundled validator)
+        run: python -m app.validate
+
+  engine-validate:
+    needs: self-validate
+    uses: GetTechAPI/TechEngine/.github/workflows/validate-data.yml@main
+    with:
+      data-ref: ${{ github.event.pull_request.head.sha || github.sha }}