Clean up state variables in MultiIndex (#16203)

MultiIndex sets it's own state variables outside of `__init__` and allows some uninitialized private variables that may be called in other methods. This PR now ensures these state variables are always initialized in `__init__`, `_from_data` and `_simple_new`

Authors:
  - Matthew Roeschke (https://github.com/mroeschke)
  - GALI PREM SAGAR (https://github.com/galipremsagar)

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

URL: #16203

python/cudf/cudf/core/dataframe.py

@@ -3593,15 +3593,15 @@ def rename(
 
             if level is not None and isinstance(self.index, MultiIndex):
                 level = self.index._get_level_label(level)
-                out_index = self.index.copy(deep=copy)
-                level_values = out_index.get_level_values(level)
-                level_values.to_frame().replace(
+                level_values = self.index.get_level_values(level)
+                ca = self.index._data.copy(deep=copy)
+                ca[level] = level_values._column.find_and_replace(
                     to_replace=list(index.keys()),
-                    value=list(index.values()),
-                    inplace=True,
+                    replacement=list(index.values()),
+                )
+                out_index = type(self.index)._from_data(
+                    ca, name=self.index.name
                 )
-                out_index._data[level] = column.as_column(level_values)
-                out_index._compute_levels_and_codes()
             else:
                 to_replace = list(index.keys())
                 vals = list(index.values())
@@ -7058,12 +7058,8 @@ def stack(self, level=-1, dropna=no_default, future_stack=False):
         # Assemble the final index
         new_index_columns = [*repeated_index._columns, *tiled_index]
         index_names = [*self.index.names, *unique_named_levels.names]
-        new_index = MultiIndex.from_frame(
-            DataFrame._from_data(
-                dict(zip(range(0, len(new_index_columns)), new_index_columns))
-            ),
-            names=index_names,
-        )
+        new_index = MultiIndex._from_data(dict(enumerate(new_index_columns)))
+        new_index.names = index_names
 
         # Compute the column indices that serves as the input for
         # `interleave_columns`