[Bug]: Pyspark reduction does not preserve names root names with nw.all()

[FBruzzesi]

Describe the bug

Pyspark reductions do not preserve original root name when applied to nw.all().

I am tempted to say that the issue is in this line. However I could not manage to fix it without breaking the code elsewhere so far 🙈

Steps or code to reproduce the bug

Running:

df.select(nw.all().n_unique())

with a lazyframe backed by pyspark DataFrame, will result in a column named (count(DISTINCT a) + max(CAST((a IS NULL) AS INT))).

Expected results

Original names to be preserved

Actual results

Pyspark expression names

Please run narwhals.show_version() and enter the output below.

Latest

Relevant log output

No response