check_sorted ignored for DataFrame.rolling when group_by is not given

[MarcoGorelli]

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

In [15]: df = pl.DataFrame({'a': [1,2,3], 'b': [date(2020, 1, 1), date(2020, 1, 2), date(2020, 1, 3)], 'c': [1, 1, 2]})

In [16]: df.rolling('b', period='2d', check_sorted=False).agg(pl.sum('a'))

Log output

---------------------------------------------------------------------------
InvalidOperationError                     Traceback (most recent call last)
Cell In[16], line 1
----> 1 df.rolling('b', period='2d').agg(pl.sum('a'), check_sorted=False)

File ~/scratch/.venv/lib/python3.11/site-packages/polars/dataframe/group_by.py:881, in RollingGroupBy.agg(self, *aggs, **named_aggs)
    852 def agg(
    853     self,
    854     *aggs: IntoExpr | Iterable[IntoExpr],
    855     **named_aggs: IntoExpr,
    856 ) -> DataFrame:
    857     """
    858     Compute aggregations for each group of a group by operation.
    859
   (...)
    868         The resulting columns will be renamed to the keyword used.
    869     """
    870     return (
    871         self.df.lazy()
    872         .rolling(
    873             index_column=self.time_column,
    874             period=self.period,
    875             offset=self.offset,
    876             closed=self.closed,
    877             by=self.by,
    878             check_sorted=self.check_sorted,
    879         )
    880         .agg(*aggs, **named_aggs)
--> 881         .collect(no_optimization=True)
    882     )

File ~/scratch/.venv/lib/python3.11/site-packages/polars/lazyframe/frame.py:1943, in LazyFrame.collect(self, type_coercion, predicate_pushdown, projection_pushdown, simplify_expression, slice_pushdown, comm_subplan_elim, comm_subexpr_elim, no_optimization, streaming, background, _eager)
   1940 if background:
   1941     return InProcessQuery(ldf.collect_concurrently())
-> 1943 return wrap_df(ldf.collect())

InvalidOperationError: argument in operation 'group_by_rolling' is not explicitly sorted

- If your data is ALREADY sorted, set the sorted flag with: '.set_sorted()'.
- If your data is NOT sorted, sort the 'expr/series/column' first.

Issue description

check_sorted should silence the error message

Expected behavior

shape: (3, 2)
┌────────────┬─────┐
│ b          ┆ a   │
│ ---        ┆ --- │
│ date       ┆ i64 │
╞════════════╪═════╡
│ 2020-01-01 ┆ 1   │
│ 2020-01-02 ┆ 3   │
│ 2020-01-03 ┆ 5   │
└────────────┴─────┘

Installed versions

--------Version info---------
Polars:               0.20.16
Index type:           UInt32
Platform:             Linux-5.15.146.1-microsoft-standard-WSL2-x86_64-with-glibc2.35
Python:               3.11.8 (main, Feb 25 2024, 16:39:33) [GCC 11.4.0]

----Optional dependencies----
adbc_driver_manager:  <not installed>
cloudpickle:          <not installed>
connectorx:           <not installed>
deltalake:            <not installed>
fastexcel:            <not installed>
fsspec:               <not installed>
gevent:               <not installed>
hvplot:               0.9.2
matplotlib:           3.8.3
numpy:                1.26.4
openpyxl:             <not installed>
pandas:               2.0.0
pyarrow:              15.0.1
pydantic:             2.6.1
pyiceberg:            <not installed>
pyxlsb:               <not installed>
sqlalchemy:           <not installed>
xlsx2csv:             <not installed>
xlsxwriter:           <not installed>

[MarcoGorelli]

Related, https://docs.pola.rs/py-polars/html/reference/expressions/api/polars.Expr.rolling.html mentions by in the check_sorted description, even though it takes no such argument

[MarcoGorelli]

Same with group_by_dynamic:

df.group_by_dynamic('b', every='2d', check_sorted=False).agg(pl.sum('a'))

(same error)