Update RateOfChange input types to only accept Datetime time index

[thehomebrewnerd]

Update RateOfChange input types to only accept Datetime time index

Fixes #2395

[codecov]

Codecov Report

Merging #2408 (1afe902) into main (3ed040d) will increase coverage by 29.55%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##             main    #2408       +/-   ##
===========================================
+ Coverage   69.89%   99.44%   +29.55%     
===========================================
  Files         326      326               
  Lines       20208    20208               
===========================================
+ Hits        14124    20096     +5972     
+ Misses       6084      112     -5972     

Impacted Files	Coverage Δ
...tives/standard/transform/numeric/rate_of_change.py	100.00% <100.00%> (ø)
featuretools/entityset/entityset.py	99.22% <0.00%> (+0.12%)	⬆️
...s/computational_backends/feature_set_calculator.py	98.69% <0.00%> (+0.78%)	⬆️
featuretools/computational_backends/feature_set.py	100.00% <0.00%> (+0.93%)	⬆️
featuretools/synthesis/deep_feature_synthesis.py	99.77% <0.00%> (+1.10%)	⬆️
featuretools/synthesis/utils.py	100.00% <0.00%> (+2.77%)	⬆️
...tandard/transform/natural_language/count_string.py	100.00% <0.00%> (+2.85%)	⬆️
...ansform/natural_language/number_of_unique_words.py	100.00% <0.00%> (+3.03%)	⬆️
featuretools/utils/time_utils.py	95.16% <0.00%> (+3.22%)	⬆️
...s/standard/transform/latlong/cityblock_distance.py	100.00% <0.00%> (+3.70%)	⬆️
... and 89 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[sbadithe]

Do we need to add a test that numeric time indexes don't work?

[thehomebrewnerd]

Do we need to add a test that numeric time indexes don't work?

I think that should already be covered by our features tests - we know DFS doesn't build features with input types that are not specified, so adding a specific test here seems unnecessary. The inclusion of all time index types in the primitive definition before was just an oversight.

[sbadithe]

Do we need to add a test that numeric time indexes don't work?

I think that should already be covered by our features tests - we know DFS doesn't build features with input types that are not specified, so adding a specific test here seems unnecessary. The inclusion of all time index types in the primitive definition before was just an oversight.

Makes sense to me!