fix: Respect nulls_last parameter in aggregate sort_by

[mcrumiller]

Fixes #17243.

From the issue example:

df = pl.DataFrame({
    "a": [1, 10, 9, 3, 2, 4, 7, 5, 6, 8, None, None],
    "b": [1, 2, 3, 1, 4, 2, 3, 3, 2, 1, 1, 2],
})

pl.Config.set_tbl_rows(12)
print(df.select(pl.all().sort_by("a", nulls_last=True).over("b")))

shape: (12, 2)
┌──────┬─────┐
│ a    ┆ b   │
│ ---  ┆ --- │
│ i64  ┆ i64 │
╞══════╪═════╡
│ 1    ┆ 1   │
│ 4    ┆ 2   │
│ 5    ┆ 3   │
│ 3    ┆ 1   │
│ 2    ┆ 4   │
│ 6    ┆ 2   │
│ 7    ┆ 3   │
│ 9    ┆ 3   │
│ 10   ┆ 2   │
│ 8    ┆ 1   │
│ null ┆ 1   │
│ null ┆ 2   │
└──────┴─────┘

[codecov]

Codecov Report

Attention: Patch coverage is 83.33333% with 1 line in your changes missing coverage. Please review.

Project coverage is 80.84%. Comparing base (01a65f0) to head (69af2c9).

Files	Patch %	Lines
crates/polars-expr/src/expressions/sortby.rs	83.33%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #17249      +/-   ##
==========================================
+ Coverage   80.82%   80.84%   +0.02%     
==========================================
  Files        1466     1466              
  Lines      192324   192328       +4     
  Branches     2745     2745              
==========================================
+ Hits       155437   155486      +49     
+ Misses      36384    36339      -45     
  Partials      503      503              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ritchie46]

Great, thanks!