Series.shift does not always preserve input dtype, but is typed as if it does

[MarcoGorelli]

Describe the bug
``

To Reproduce

1. Provide a minimal runnable pandas example that is not properly checked by the stubs.
2. Indicate which type checker you are using (mypy or pyright).
3. Show the error message received from that type checker while checking your example.

import pandas as pd


reveal_type(pd.Series([1,2,3]).shift())
#  t.py:4: note: Revealed type is "pandas.core.series.Series[builtins.int]"

however, the result is actually Series['float']

It's not clear to me if/how pandas-stubs distinguishes ints and nullable ints #1110

Please complete the following information:

• OS: [e.g. Windows, Linux, MacOS] linux
• OS Version [e.g. 22] 22.04
• python version 3.12.5
• version of type checker mypy 1.15
• version of installed pandas-stubs 2.2.3.241126

Additional context
noticed looking into #1108

[MarcoGorelli]

This one might quite complex to type correctly?

• shift(n) changes from 'int' to 'float' (and from 'bool' to 'object'), unless:
  • n is 0
  • fill_value is specified as a compatible type
  • freq is also specified

[Dr-Irv]

As I stated here: #1108 (comment) , I don't think we can do this because the value of freq and the underlying index could change the result.

I think we have to just return Series, since we can't infer the subtype.