BUG: Fix alignment of list-like objects for operations with DataFrame

[jschendel]

• closes Wrong orientation of operations between DataFrame and range() #17901
• tests added / passed
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

Some comments:

• Replaced pd.DataFrame with ABCDataFrame in isinstance checks throughout ops.py, as I'm of the impression that this is the best practice. Can revert these if I'm mistaken.
• I didn't use deque in the test I modified (though I did in the one I created). The reason being that deque gets cast to int32 instead of int64 on Windows, and the test compares against a Series with int64 dtype.
  • Could be fixed by mimicking the is_list_like check here, and doing something similar to BUG: Fix range dtype in Series/DataFrame constructor on Windows #17840 but probably best to do in a separate PR (would be happy to do such a PR)
  • I could explicitly cast the result series to int64 if we want to add deque now, then remove after implementing a fix
  • I suppose it could be marked as xfailing if without modifications if we want to add deque now, but seems kind of pointless to mark one of the tests being modified as xfailing

[codecov]

Codecov Report

Merging #17926 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #17926      +/-   ##
==========================================
- Coverage   91.23%   91.22%   -0.02%     
==========================================
  Files         163      163              
  Lines       50112    50113       +1     
==========================================
- Hits        45722    45714       -8     
- Misses       4390     4399       +9

Flag	Coverage Δ
#multiple	89.03% <100%> (ø)	⬆️
#single	40.31% <70%> (-0.07%)	⬇️

Impacted Files	Coverage Δ
pandas/core/ops.py	91.77% <100%> (+0.01%)	⬆️
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/core/frame.py	97.75% <0%> (-0.1%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 51c5f4d...782eddc. Read the comment docs.

[jreback]

thanks @jschendel nice fix!

I didn't use deque in the test I modified (though I did in the one I created). The reason being that deque gets cast to int32 instead of int64 on Windows, and the test compares against a Series with int64 dtype.

FYI you can construct using np.arange() which is platform dependent.