Remove Dask and Spark DataFrame Support

[thehomebrewnerd]

Closes #2704

Removes support for creating EntitySets from Dask and Spark dataframes.

[codecov]

Codecov Report

Attention: Patch coverage is 99.85141% with 1 lines in your changes are missing coverage. Please review.

Project coverage is 99.38%. Comparing base (21d0bf0) to head (dcee907).

Files	Patch %	Lines
...s/computational_backends/feature_set_calculator.py	87.50%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #2705       +/-   ##
===========================================
+ Coverage   86.97%   99.38%   +12.41%     
===========================================
  Files         404      397        -7     
  Lines       24230    22292     -1938     
===========================================
+ Hits        21073    22156     +1083     
+ Misses       3157      136     -3021     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[tamargrey]

A couple places that probably still need to be updated:

• __dask_tokenize__ still exists in entityset.py - probably can be removed
• The Parallel Computation by Partitioning Data section in the performance.ipynb file has some references to spark - Not sure if we want to just remove all references entirely or leave some subset of it (maybe just reference that an old version of featuretools can be used if you want to use it with Spark), but we should definitely remove the link to the Feature Engineering on Spark Notebook, which will no longer exist.

[thehomebrewnerd]

• The Parallel Computation by Partitioning Data section in the performance.ipynb file has some references to spark - Not sure if we want to just remove all references entirely or leave some subset of it (maybe just reference that an old version of featuretools can be used if you want to use it with Spark), but we should definitely remove the link to the Feature Engineering on Spark Notebook, which will no longer exist.

@tamargrey I think those references are still valid since they relate to manually partitioning data and not doing it via pyspark dataframes in an EntitySet. The Feature Engineering on Spark Notebook will still exist here: https://github.com/alteryx/predict-customer-churn/blob/main/churn/4.%20Feature%20Engineering%20on%20Spark.ipynb. The linked article in that section is gone though, so I'll remove that reference.

[tamargrey]

lgtm!