MAINT: adapt to more general scipy warnings

more specific warnings were removed for scipy 1.9.0.
see https://docs.scipy.org/doc/scipy/release.1.9.0.html#backwards-incompatible-changes

Author: gregcaporaso

setup.py

@@ -152,7 +152,7 @@
           'natsort >= 4.0.3',
           'numpy >= 1.9.2',
           'pandas >= 1.0.0',
-          'scipy >= 1.3.0',
+          'scipy >= 1.9.0',
           'h5py >= 2.9.0',
           'hdmedians >= 0.14.1',
           'scikit-learn >= 0.19.1',

skbio/stats/distance/_mantel.py

@@ -13,9 +13,9 @@
 import pandas as pd
 import scipy.special
 from scipy.stats import kendalltau
-from scipy.stats import PearsonRConstantInputWarning
-from scipy.stats import PearsonRNearConstantInputWarning
-from scipy.stats import SpearmanRConstantInputWarning
+from scipy.stats import ConstantInputWarning
+from scipy.stats import NearConstantInputWarning
+from scipy.stats import ConstantInputWarning
 
 from skbio.stats.distance import DistanceMatrix
 from skbio.util._decorator import experimental
@@ -352,7 +352,7 @@ def _mantel_stats_pearson_flat(x, y_flat, permutations):
 
     # If an input is constant, the correlation coefficient is not defined.
     if (x_flat == x_flat[0]).all() or (y_flat == y_flat[0]).all():
-        warnings.warn(PearsonRConstantInputWarning())
+        warnings.warn(ConstantInputWarning())
         return np.nan, np.nan, []
 
     # inline pearsonr, condensed from scipy.stats.pearsonr
@@ -375,7 +375,7 @@ def _mantel_stats_pearson_flat(x, y_flat, permutations):
         # If all the values in x (likewise y) are very close to the mean,
         # the loss of precision that occurs in the subtraction xm = x - xmean
         # might result in large errors in r.
-        warnings.warn(PearsonRNearConstantInputWarning())
+        warnings.warn(NearConstantInputWarning())
 
     orig_stat = np.dot(xm_normalized, ym_normalized)
 
@@ -471,7 +471,7 @@ def _mantel_stats_spearman(x, y, permutations):
 
     # If an input is constant, the correlation coefficient is not defined.
     if (x_flat == x_flat[0]).all() or (y_flat == y_flat[0]).all():
-        warnings.warn(SpearmanRConstantInputWarning())
+        warnings.warn(ConstantInputWarning())
         return np.nan, np.nan, []
 
     y_rank = scipy.stats.rankdata(y_flat)