Confusing failures using kwarg-based init of dataframes about sciris HOT 1 CLOSED

import pandas as pd
pd.DataFrame(data={'a':[1,2,3]}, columns=[4,5,6])
Out[13]: 
Empty DataFrame
Columns: [4, 5, 6]
Index: []

sc.dataframe(a=[1, 2, 3], cols=['a','b'])
Out[3]: 
   a    b
0  1  NaN
1  2  NaN
2  3  NaN

sc.dataframe(a=[1, 2, 3], cols=[4, 5, 6])
/tmp/ipykernel_592135/2569439549.py:1: RuntimeWarning: No overlap between column names and data keys, are you sure you want to do this?
Columns: {4, 5, 6}
Data keys: {'a'}

Out[2]: 
Empty dataframe
Columns: [4, 5, 6]
Index: []

sc.dataframe(a=[1, 2, 3], data=[4, 5, 6])
Traceback (most recent call last):

  Cell In[1], line 1
    sc.dataframe(a=[1, 2, 3], data=[4, 5, 6])

  File ~/sciris/sciris/sciris/sc_dataframe.py:101 in __init__
    raise TypeError(errormsg)

TypeError: When providing data columns via keywords ("a"), these can only be combined with a dict, not an object of <class 'list'>. Pass the data as a dict instead.