Early crash fix

[perib]

[please review the Contribution Guidelines prior to submitting your pull request. go ahead and delete this line if you've already reviewed said guidelines.]

What does this PR do?

I believe this addresses the cause of the issue described in #1313 and #1308.

The issue in in line 134 and 137 of gp_deap.py

Original

if ind1 is not None:
                ind1, _ = toolbox.mate(ind1, ind2)
                del ind1.fitness.values

bugged

if ind1 is not None:
                ind1_cx, _, evaluated_individuals_= toolbox.mate(ind1, ind2)
                del ind1.fitness.values

fixed

if ind1 is not None:
                ind1_cx, _, evaluated_individuals_= toolbox.mate(ind1, ind2)
                del ind1_cx.fitness.values

In commit 9e6ad7e line 134 was changed to rename ind1 to ind1_cx.
However the following line was not changed to rename ind1 to ind1_cx. So rather than deleting the values for the new individual, it was deleting the scores for the old individual. likely triggering the error identified by @Lingepumpe in #1308

How should this PR be tested?

I used the following code to reproduce the behavior. It doesn't happen all of the time, needs to be rerun a few times to recreate the problem. The issue does not seem to happen with the fixed version.

from tpot import TPOTRegressor
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
import pandas as pd
import numpy as np
import sklearn.datasets

X, y = sklearn.datasets.make_regression(n_samples=100)

est = TPOTRegressor(generations=50, population_size=28, cv=2, verbosity=2, random_state=42, n_jobs=-2)

est.fit(X,y)

What are the relevant issues?

#1313 #1308