Fix dataclass definition error in AutoEvalColumn

- Replace problematic make_dataclass usage with simple class definition
- Fix 'mutable default ColumnContent for field rank is not allowed' error
- This resolves the issue where clicking 'About' tab causes space to stop responding
- All components now initialize successfully

src/display/utils.py

@@ -23,19 +23,16 @@ class ColumnContent:
 
 
 ## Leaderboard columns
-auto_eval_column_dict = []
-# Init
-auto_eval_column_dict.append(["rank", ColumnContent, ColumnContent("Rank", "number", True, never_hidden=True)])
-auto_eval_column_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
-# Scores
-auto_eval_column_dict.append(["average", ColumnContent, ColumnContent("Average Score", "number", True)])
+# Create a simple class to hold column definitions
+class AutoEvalColumn:
+    rank = ColumnContent("Rank", "number", True, never_hidden=True)
+    model = ColumnContent("Model", "markdown", True, never_hidden=True)
+    average = ColumnContent("Average Score", "number", True)
+    model_size = ColumnContent("Model Size", "str", True)
+
+# Add task columns dynamically
 for task in Tasks:
-    auto_eval_column_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
-# Model information
-auto_eval_column_dict.append(["model_size", ColumnContent, ColumnContent("Model Size", "str", True)])
-
-# We use make dataclass to dynamically fill the scores from Tasks
-AutoEvalColumn = make_dataclass("AutoEvalColumn", auto_eval_column_dict, frozen=True)
+    setattr(AutoEvalColumn, task.name, ColumnContent(task.value.col_name, "number", True))
 
 
 ## For the queue columns in the submission tab