Allow passing entities without relationships to CFM

[rwedge]

Fixes #1287

Before, calculate_feature_matrix would check that bothentities and relationships were not None before creating an entityset with them, which required passing an empty iterable of relationships in order to work with a single entity. This PR changes that by removing that check and relying on the EntitySet init method to handle valdiating the entities and relationships

[codecov]

Codecov Report

Merging #1290 (91e347d) into main (723662f) will increase coverage by 0.01%.
The diff coverage is 97.91%.

@@            Coverage Diff             @@
##             main    #1290      +/-   ##
==========================================
+ Coverage   98.36%   98.38%   +0.01%     
==========================================
  Files         134      134              
  Lines       14456    14473      +17     
==========================================
+ Hits        14220    14239      +19     
+ Misses        236      234       -2     

Impacted Files	Coverage Δ
featuretools/tests/synthesis/test_dfs_method.py	100.00% <ø> (+0.44%)	⬆️
featuretools/tests/conftest.py	96.59% <96.55%> (-0.02%)	⬇️
...computational_backends/calculate_feature_matrix.py	99.70% <100.00%> (+0.59%)	⬆️
...utational_backend/test_calculate_feature_matrix.py	99.44% <100.00%> (+0.01%)	⬆️

---

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 723662f...91e347d. Read the comment docs.

[tamargrey]

Just wondering--why not keep the check for whether entities is not None here? Is there a world in which having no entities wouldn't end up erroring?

If not, it might be nice to catch this with its own error instead of Entity transactions does not exist in entityset, which is where it gets caught in the test below.

[rwedge]

Sure, if we catch there we can give a more descriptive error

[tamargrey]

lgtm!