perf: optimize DataFrame.describe by presorting columns

By presorting numerical columns, quantiles/min/max will be O(1)

py-polars/polars/dataframe/frame.py

@@ -4451,8 +4451,14 @@ def describe(
             for c in self.columns
         ]
 
+        # Sort numerical columns to make all quantiles O(1)
+        sort_exprs = [
+            (F.col(c).sort() if c in stat_cols else F.col(c)) for c in self.columns
+        ]
+        df_cols_presorted = self.select(*sort_exprs)
+
         # Calculate metrics in parallel
-        df_metrics = self.select(
+        df_metrics = df_cols_presorted.select(
             F.all().count().name.prefix("count:"),
             F.all().null_count().name.prefix("null_count:"),
             *mean_exprs,