Add proper support for tolerances in testing methods. (#8649)

Resolves #8646 so that testing equality between different types of frames can be based on approximate rather than exact equality. Note that this is a blocker for packages that need to move away from relying on `cudf.tests.utils` for testing functions, since that module is no longer exposed by `cudf`.

Authors:
  - Vyas Ramasubramani (https://github.com/vyasr)

Approvers:
  - https://github.com/brandon-b-miller
  - GALI PREM SAGAR (https://github.com/galipremsagar)

URL: #8649

python/cudf/cudf/testing/testing.py

@@ -4,10 +4,12 @@
 
 from typing import Union
 
+import cupy as cp
 import numpy as np
 import pandas as pd
 
 import cudf
+from cudf.api.types import is_numeric_dtype
 from cudf.core._compat import PANDAS_GE_110
 from cudf.utils.dtypes import is_categorical_dtype
 
@@ -203,7 +205,17 @@ def assert_column_equal(
 
     columns_equal = False
     try:
-        columns_equal = left.equals(right)
+        columns_equal = (
+            (
+                cp.all(left.isnull().values == right.isnull().values)
+                and cp.allclose(
+                    left[left.isnull().unary_operator("not")].values,
+                    right[right.isnull().unary_operator("not")].values,
+                )
+            )
+            if not check_exact and is_numeric_dtype(left)
+            else left.equals(right)
+        )
     except TypeError as e:
         if str(e) != "Categoricals can only compare with the same type":
             raise e