Add any-to-any task category and language tags

Hi! I'm Niels from the Hugging Face community science team.

This PR updates the dataset card metadata to include the `any-to-any` task category and language tags. These additions help researchers and users find your dataset more easily when filtering for multimodal reasoning benchmarks on the Hub.

The rest of the dataset card looks great, providing clear instructions, a leaderboard, and proper citation. Thanks for documenting this work!

README.md

@@ -1,4 +1,8 @@
 ---
+task_categories:
+- any-to-any
+language:
+- en
 configs:
 - config_name: main
   data_files:
@@ -16,9 +20,12 @@ configs:
     path: paperfolding/*.parquet
   - split: sokoban
     path: sokoban/*.parquet
+tags:
+- multimodal
+- reasoning
+- world-models
 ---
 
-
 # VisWorld-Eval: Task Suite for Reasoning with Visual World Modeling 🌏
 
 [![Project Page](https://img.shields.io/badge/Project_Page-blue)](https://thuml.github.io/Reasoning-Visual-World/)