rank() on a Series of just 1 null assigns rank=1 to the null value.

[rgiaccio]

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

import polars as pl

print(pl.Series([]).rank().to_list())
print(pl.Series([None]).rank().to_list())
print(pl.Series([None, None]).rank().to_list())
print(pl.Series([None, None, None]).rank().to_list())

Output:

[]
[1.0]
[None, None]
[None, None, None]

Log output

No response

Issue description

Ranks of series [null] should not be [1.0] but [null], because the rank of null should be null.
This only happens with series of just 1 null. Ranking series with 0 or more than 1 null gives correct results.

This issue is a generalization of #17570

Expected behavior

import polars as pl

print(pl.Series([]).rank().to_list())
print(pl.Series([None]).rank().to_list())
print(pl.Series([None, None]).rank().to_list())
print(pl.Series([None, None, None]).rank().to_list())

Output:

[]
[None]
[None, None]
[None, None, None]

Installed versions

--------Version info---------
Polars:               1.5.0
Index type:           UInt32
Platform:             macOS-14.5-x86_64-i386-64bit
Python:               3.11.4 (main, Oct  4 2023, 02:12:57) [Clang 15.0.0 (clang-1500.0.40.1)]

----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>
great_tables:         <not installed>
hvplot:               <not installed>
matplotlib:           <not installed>
nest_asyncio:         1.6.0
numpy:                2.0.1
openpyxl:             <not installed>
pandas:               <not installed>
pyarrow:              <not installed>
pydantic:             <not installed>
pyiceberg:            <not installed>
sqlalchemy:           <not installed>
torch:                <not installed>
xlsx2csv:             <not installed>
xlsxwriter:           <not installed>

[ritchie46]

Thanks for the simplification!

[rgiaccio]

Thanks for the great work!