Series.to_numpy raises when zero_copy_only=True even when Series is empty

[stinodego]

Checks

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest version of Polars.

Reproducible example

s = pl.Series()
arr = s.to_numpy(use_pyarrow=False, zero_copy_only=True)

Log output

Traceback (most recent call last):
  File "/home/stijn/code/polars/py-polars/repro.py", line 21, in <module>
    arr2 = s.to_numpy(use_pyarrow=False, zero_copy_only=True)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/stijn/code/polars/py-polars/polars/series/series.py", line 4397, in to_numpy
    raise_no_zero_copy()
  File "/home/stijn/code/polars/py-polars/polars/series/series.py", line 4344, in raise_no_zero_copy
    raise ValueError(msg)
ValueError: cannot return a zero-copy array

Issue description

I would think this should not raise, since we're not copying any data.

PyArrow also raises here though:

s = pl.Series()
arr = s.to_numpy(use_pyarrow=True, zero_copy_only=True)
# pyarrow.lib.ArrowInvalid: Needed to copy 1 chunks with 0 nulls, but zero_copy_only was True

Expected behavior

Same behavior as when zero_copy_only=False.

Installed versions

main