Fix groupby aggregation to return plain DataFrame without spurious weights

Aggregated results from groupby operations should be plain DataFrames,
not MicroDataFrames, since the weights have already been applied during
aggregation. This fixes the NaN weight column issue noted in review.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

Author: MaxGhenis

microdf/microdataframe.py

@@ -765,11 +765,9 @@ def fn(*args, **kwargs):
                         except Exception:
                             # Skip columns that can't be aggregated
                             pass
-                    return (
-                        MicroDataFrame(results)
-                        if results
-                        else MicroDataFrame()
-                    )
+                    # Return plain DataFrame - aggregated results don't have
+                    # per-row weights (weights were already applied)
+                    return pd.DataFrame(results) if results else pd.DataFrame()
 
                 return fn
 
@@ -787,11 +785,9 @@ def fn(*args, **kwargs) -> Union[pd.Series, pd.DataFrame]:
                         except Exception:
                             # Skip columns that can't be aggregated
                             pass
-                    return (
-                        MicroDataFrame(results)
-                        if results
-                        else MicroDataFrame()
-                    )
+                    # Return plain DataFrame - aggregated results don't have
+                    # per-row weights (weights were already applied)
+                    return pd.DataFrame(results) if results else pd.DataFrame()
 
                 return fn
 
@@ -850,10 +846,12 @@ def fn(*args, **kwargs):
                                 )(*args, **kwargs)
                             except Exception:
                                 pass
+                        # Return plain DataFrame - aggregated results don't
+                        # have per-row weights (weights were already applied)
                         return (
-                            MicroDataFrame(results)
+                            pd.DataFrame(results)
                             if results
-                            else MicroDataFrame()
+                            else pd.DataFrame()
                         )
 
                     return fn
@@ -871,10 +869,12 @@ def fn(*args, **kwargs):
                                 )(*args, **kwargs)
                             except Exception:
                                 pass
+                        # Return plain DataFrame - aggregated results don't
+                        # have per-row weights (weights were already applied)
                         return (
-                            MicroDataFrame(results)
+                            pd.DataFrame(results)
                             if results
-                            else MicroDataFrame()
+                            else pd.DataFrame()
                         )
 
                     return fn

microdf/tests/test_microseries_dataframe.py

@@ -339,3 +339,8 @@ def test_groupby_column_selection() -> None:
     result_list = d.groupby("g")[["y"]].sum()
     assert result_list.loc["a", "y"] == 14.0
     assert result_list.loc["b", "y"] == 18.0
+
+    # Aggregated results should be plain DataFrame (no spurious weight column)
+    result_all = d.groupby("g").sum()
+    assert "weight" not in result_all.columns
+    assert list(result_all.columns) == ["y"]