Nested struct column is null after pivoting DataFrame about polars

[cmdlineluser]

Can reproduce.

In case it may be a useful datapoint for debugging, transpose also seems to lose the inner values.

>>> df.drop('foo').transpose(column_names='bar')
# shape: (1, 4)
# ┌────────────┬────────────┬────────────┬────────────┐
# │ x          ┆ x          ┆ y          ┆ y          │
# │ ---        ┆ ---        ┆ ---        ┆ ---        │
# │ struct[2]  ┆ struct[2]  ┆ struct[2]  ┆ struct[2]  │
# ╞════════════╪════════════╪════════════╪════════════╡
# │ {1,{null}} ┆ {3,{null}} ┆ {5,{null}} ┆ {7,{null}} │
# └────────────┴────────────┴────────────┴────────────┘