remove convert_links_to_integers

[Seth-Rothschild]

Removes the keyword argument convert_links_to_integers from normalize entity.

Some of the last_time_index tests were using a normalization with convert_links_to_integers and sorting values, leading to a resort of the "true" values in the test.

Original ordering was

0.0, 5.0, 10.0, 15.0, 20.0, 1.0, 2.0, 3.0, 7.0, 14.0, NaN

if we convert these to range(11) and sort them according to the value in the list above, the new ordering of that range would become

0, 5, 6, 7, 1, 8, 2, 9, 3, 4, 10

which aligns exactly with the changes we had to make in true_values_lti.

The other change, changing the value in test_parent_some_missing, is really a convenience measure to ensure that it ends up second to last on the sorted list. If we put it earlier, we'll have to move all the rows instead of just one.

[codecov-io]

Codecov Report

Merging #219 into master will decrease coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #219      +/-   ##
==========================================
- Coverage   93.53%   93.52%   -0.01%     
==========================================
  Files          71       71              
  Lines        7759     7749      -10     
==========================================
- Hits         7257     7247      -10     
  Misses        502      502

Impacted Files	Coverage Δ
featuretools/entityset/entityset.py	94.3% <ø> (-0.3%)	⬇️
...ools/tests/entityset_tests/test_last_time_index.py	100% <100%> (ø)	⬆️
featuretools/tests/entityset_tests/test_es.py	99.31% <100%> (ø)	⬆️
featuretools/utils/gen_utils.py	66.66% <0%> (+2.22%)	⬆️

---

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 f74e7c3...aca37ae. Read the comment docs.

[kmax12]

This looks good to me!