docs: add v2 config split (130 tasks) to dataset frontmatter

README.md

@@ -29,7 +29,11 @@ configs:
   - config_name: default
     data_files:
       - split: test
-        path: data/train-00000-of-00001.parquet
+        path: data/train-*.parquet
+  - config_name: v2
+    data_files:
+      - split: test
+        path: data/v2-*.parquet
 arxiv: "2604.08523"
 viewer: true
 leaderboard: TIGER-Lab/ClawBench