BUG: groupby with dropna=False and pa.dictionary drops NA values

[rhshadrach]

df = pd.DataFrame({'A': ['a1', pd.NA]}, dtype=pd.ArrowDtype(pa.dictionary(pa.int32(), pa.utf8())))
print(df.groupby("A", dropna=False)[["A"]].first())
#      A
# A     
# a1  a1

There should be a 2nd row with the NA value since dropna=False.