Drop use of six

This package has an undeclared requirement on six, but since Python 2 is
long unsupported, remove it and just use the Python 3 idiom.

Author: s-t-e-v-e-n-k

sklearn_pandas/pipeline.py

@@ -1,4 +1,3 @@
-import six
 from sklearn.pipeline import _name_estimators, Pipeline
 from sklearn.utils import tosequence
 
@@ -59,7 +58,7 @@ def __init__(self, steps):
 
     def _pre_transform(self, X, y=None, **fit_params):
         fit_params_steps = dict((step, {}) for step, _ in self.steps)
-        for pname, pval in six.iteritems(fit_params):
+        for pname, pval in fit_params.items():
             step, param = pname.split('__', 1)
             fit_params_steps[step][param] = pval
         Xt = X