Update argument order of objectives to align with sklearn

[angela97lin]

Closes #662

[codecov]

Codecov Report

Merging #698 into master will not change coverage.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #698   +/-   ##
=======================================
  Coverage   99.21%   99.21%           
=======================================
  Files         140      140           
  Lines        4985     4985           
=======================================
  Hits         4946     4946           
  Misses         39       39           

Impacted Files	Coverage Δ
...alml/objectives/binary_classification_objective.py	100.00% <100.00%> (ø)
evalml/objectives/fraud_cost.py	100.00% <100.00%> (ø)
evalml/objectives/lead_scoring.py	95.00% <100.00%> (ø)
evalml/objectives/objective_base.py	100.00% <100.00%> (ø)
evalml/objectives/standard_metrics.py	100.00% <100.00%> (ø)
evalml/pipelines/binary_classification_pipeline.py	100.00% <100.00%> (ø)
...ml/pipelines/multiclass_classification_pipeline.py	100.00% <100.00%> (ø)
evalml/pipelines/pipeline_base.py	99.51% <100.00%> (ø)
evalml/pipelines/plot_utils.py	100.00% <100.00%> (ø)
...alml/tests/objective_tests/test_fraud_detection.py	100.00% <100.00%> (ø)
... and 3 more

---

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 fafb0e8...6c29b8b. Read the comment docs.

[dsherry]

The changes look good.

Because this is a dangerous change to make, we should triple-check it. Once you merge #707 and update this PR, ping me and I'll check the branch out and we should each poke around to ensure we didn't miss any spots.

There weren't any callsites in the docs which needed to be updated?

[angela97lin]

@dsherry I didn't find anywhere in docs to update except our custom objectives notebook... which I realized was really outdated, yikes! I ended up just replacing it with our new updated version of FraudCost (since before, it was just a copied version of the old implementation lol), thus updating the docs with score!

[angela97lin]

@dsherry Since I removed the cleanup pipeline code, codecov no longer is complaining and thus this issue is no longer blocked by #707!

[dsherry]

@angela97lin could you please also update this one callsite in confusion_matrix? It's code I added recently. The arg order doesn't matter for the sklearn function being called but it would be great to catch all usages of the old orderings in this PR.

[dsherry]

Some other places to change: git grep -n "y_pred[^ ]* y_true" prints out:

# i already mentioned this one above
evalml/pipelines/plot_utils.py:33:    labels = unique_labels(y_predicted, y_true)
evalml/tests/objective_tests/test_standard_metrics.py:37:        obj.score(y_predicted=[], y_true=[1])
evalml/tests/objective_tests/test_standard_metrics.py:39:        obj.score(y_predicted=[1], y_true=[])
evalml/tests/objective_tests/test_standard_metrics.py:41:        obj.score(y_predicted=[0], y_true=[1, 0])
evalml/tests/objective_tests/test_standard_metrics.py:43:        obj.score(y_predicted=np.array([0]), y_true=np.array([1, 0]))
evalml/tests/objective_tests/test_standard_metrics.py:60:        obj.score(y_predicted=[], y_true=[1])
evalml/tests/objective_tests/test_standard_metrics.py:62:        obj.score(y_predicted=[1], y_true=[])
evalml/tests/objective_tests/test_standard_metrics.py:64:        obj.score(y_predicted=[0], y_true=[1, 0])
evalml/tests/objective_tests/test_standard_metrics.py:66:        obj.score(y_predicted=np.array([0]), y_true=np.array([1, 0]))

Other than that, this looks good! I'll approve once updated.

[angela97lin]

@dsherry Whoops, must have missed since it was functionally equivalent (named parameters); updated!

[dsherry]

Great to get this out of the way! 🎆

Let's wait to merge #716 first and then green tests before merging