Fix: Make SimpleClassificationPipeline tests deterministic

[eddiebergman]

Fixes some nondeterminism in testing SimpleClassificationPipeline as a whole, does not change the non-deterministic of testing randomly sampled configurations.

Failing Test: https://github.com/automl/auto-sklearn/runs/4767926218?check_suite_focus=true

[codecov]

Codecov Report

Merging #1366 (fd674bb) into development (b01c732) will increase coverage by 0.00%.
The diff coverage is n/a.

@@             Coverage Diff              @@
##           development    #1366   +/-   ##
============================================
  Coverage        87.97%   87.98%           
============================================
  Files              140      140           
  Lines            10993    10993           
============================================
+ Hits              9671     9672    +1     
+ Misses            1322     1321    -1     

Impacted Files	Coverage Δ
...ine/components/classification/gradient_boosting.py	92.98% <0.00%> (-0.88%)	⬇️
autosklearn/evaluation/abstract_evaluator.py	92.27% <0.00%> (-0.78%)	⬇️
...eline/components/feature_preprocessing/fast_ica.py	97.82% <0.00%> (+6.52%)	⬆️
...ponents/feature_preprocessing/select_percentile.py	92.30% <0.00%> (+7.69%)	⬆️

---

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 b01c732...fd674bb. Read the comment docs.

[eddiebergman]

Appears failing tests are due to the MLP classifier being non-deterministic. Will handle these in a seperate PR.