Remove deprecated method DataFrame.hash_columns. (#9943)

This PR removes the deprecated method `DataFrame.hash_columns`. Users can replace existing calls like `df.hash_columns(columns, method)` with `df[columns].hash_values(method)`. Resolves #9503, follows up on #9458.

Authors:
  - Bradley Dice (https://github.com/bdice)

Approvers:
  - https://github.com/brandon-b-miller

URL: #9943

docs/cudf/source/api_docs/dataframe.rst

@@ -254,7 +254,6 @@ Serialization / IO / conversion
    DataFrame.from_arrow
    DataFrame.from_pandas
    DataFrame.from_records
-   DataFrame.hash_columns
    DataFrame.hash_values
    DataFrame.to_arrow
    DataFrame.to_dlpack

python/cudf/cudf/core/dataframe.py

@@ -4150,38 +4150,6 @@ def apply_chunks(
             tpb=tpb,
         )
 
-    def hash_columns(self, columns=None, method="murmur3"):
-        """Hash the given *columns* and return a new device array
-
-        This method is deprecated. Replace ``df.hash_columns(columns, method)``
-        with ``df[columns].hash_values(method)``.
-
-        Parameters
-        ----------
-        columns : sequence of str; optional
-            Sequence of column names. If columns is *None* (unspecified),
-            all columns in the frame are used.
-        method : {'murmur3', 'md5'}, default 'murmur3'
-            Hash function to use:
-            * murmur3: MurmurHash3 hash function.
-            * md5: MD5 hash function.
-
-        Returns
-        -------
-        Series
-            Hash values for each row.
-        """
-        warnings.warn(
-            "The `hash_columns` method will be removed in a future cuDF "
-            "release. Replace `df.hash_columns(columns, method)` with "
-            "`df[columns].hash_values(method)`.",
-            FutureWarning,
-        )
-        if columns is None:
-            # Slice by [:] to keep all columns.
-            columns = slice(None, None, None)
-        return self[columns].hash_values(method=method)
-
     def hash_values(self, method="murmur3"):
         """Compute the hash of values in each row.
 

python/cudf/cudf/tests/test_dataframe.py

@@ -1107,34 +1107,6 @@ def test_assign():
     np.testing.assert_equal(gdf2.y.to_numpy(), [2, 3, 4])
 
 
-@pytest.mark.parametrize("nrows", [1, 8, 100, 1000])
-@pytest.mark.parametrize("method", ["murmur3", "md5"])
-def test_dataframe_hash_columns(nrows, method):
-    gdf = cudf.DataFrame()
-    data = np.asarray(range(nrows))
-    data[0] = data[-1]  # make first and last the same
-    gdf["a"] = data
-    gdf["b"] = gdf.a + 100
-    with pytest.warns(FutureWarning):
-        out = gdf.hash_columns(["a", "b"])
-    assert isinstance(out, cudf.Series)
-    assert len(out) == nrows
-    assert out.dtype == np.int32
-
-    # Check default
-    with pytest.warns(FutureWarning):
-        out_all = gdf.hash_columns()
-    assert_eq(out, out_all)
-
-    # Check single column
-    with pytest.warns(FutureWarning):
-        out_one = gdf.hash_columns(["a"], method=method)
-    # First matches last
-    assert out_one.iloc[0] == out_one.iloc[-1]
-    # Equivalent to the cudf.Series.hash_values()
-    assert_eq(gdf["a"].hash_values(method=method), out_one)
-
-
 @pytest.mark.parametrize("nrows", [1, 8, 100, 1000])
 @pytest.mark.parametrize("method", ["murmur3", "md5"])
 def test_dataframe_hash_values(nrows, method):