Pandas BUG: Transpose changes dtype from object to timedelta/datetime

Pandas version checks

• [X] I have checked that this issue has not already been reported.

• [X] I have confirmed this bug exists on the latest version of pandas.

• [X] I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

df = DataFrame({"a": [pd.Timedelta("1 days"), pd.NaT, pd.Timedelta("2 days")]}, dtype=object)
df.T.dtypes


0    timedelta64[ns]
1     datetime64[ns]
2    timedelta64[ns]
dtype: object

Issue Description

I'd expected this to return object, but this was changed in #26285

Expected Behavior

The whatsnew looks like the fix was tailored to avoid raising an error, any other motivations?

Installed Versions

Replace this line with the output of pd.show_versions()

Comment From: jbrockmendel

I'd expect object too