Add WBench evaluation results

This PR adds this model's results on [**WBench**](https://github.com/meituan-longcat/WBench) — a multi-turn benchmark for interactive video world models (289 cases, 22 metrics, 5 dimensions), now registered as a benchmark on the Hub.

Scores link to the [leaderboard](https://meituan-longcat.github.io/WBench/). Feel free to close if you'd prefer not to include it — thanks for the great work!

.eval_results/wbench.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: meituan-longcat/WBench
+    task_id: wbench_navi
+  value: 64.0
+  source:
+    url: https://meituan-longcat.github.io/WBench/
+    name: WBench Leaderboard