BUG: Fix issue with negative labels in group_cumsum (#58984)

* BUG: Fix issue with negative labels in group_cumsum function

* Remove blank line

* Revert remove blank line

* Add test

* Add what's new

* typo

Author: luke396

doc/source/whatsnew/v3.0.0.rst

@@ -571,6 +571,7 @@ Groupby/resample/rolling
 - Bug in :meth:`DataFrameGroupBy.agg` that raises ``AttributeError`` when there is dictionary input and duplicated columns, instead of returning a DataFrame with the aggregation of all duplicate columns. (:issue:`55041`)
 - Bug in :meth:`DataFrameGroupBy.apply` that was returning a completely empty DataFrame when all return values of ``func`` were ``None`` instead of returning an empty DataFrame with the original columns and dtypes. (:issue:`57775`)
 - Bug in :meth:`DataFrameGroupBy.apply` with ``as_index=False`` that was returning :class:`MultiIndex` instead of returning :class:`Index`. (:issue:`58291`)
+- Bug in :meth:`DataFrameGroupBy.cumsum` where it did not return the correct dtype when the label contained ``None``. (:issue:`58811`)
 - Bug in :meth:`DataFrameGroupby.transform` and :meth:`SeriesGroupby.transform` with a reducer and ``observed=False`` that coerces dtype to float when there are unobserved categories. (:issue:`55326`)
 - Bug in :meth:`Rolling.apply` where the applied function could be called on fewer than ``min_period`` periods if ``method="table"``. (:issue:`58868`)
 

pandas/_libs/groupby.pyx

@@ -399,7 +399,12 @@ def group_cumsum(
             lab = labels[i]
 
             if lab < 0:
+                # GH#58811
+                if uses_mask:
+                    result_mask[i, :] = True
+                    out[i, :] = 0
                 continue
+
             for j in range(K):
                 val = values[i, j]
 

pandas/tests/groupby/transform/test_transform.py

@@ -1591,3 +1591,12 @@ def test_min_one_dim_no_type_coercion():
 
     expected = DataFrame({"Y": [9435, -5465765, -5465765, 0, 9435]}, dtype="int32")
     tm.assert_frame_equal(expected, result)
+
+
+def test_nan_in_cumsum_group_label():
+    # GH#58811
+    df = DataFrame({"A": [1, None], "B": [2, 3]}, dtype="Int16")
+    gb = df.groupby("A")["B"]
+    result = gb.cumsum()
+    expected = Series([2, None], dtype="Int16", name="B")
+    tm.assert_series_equal(expected, result)