[PyTorch] Don't set FP8 data to `None` when saving base tensors (#1548)

Don't set data to null
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

transformer_engine/pytorch/tensor/_internal/float8_tensor_base.py

@@ -105,8 +105,6 @@ class Float8TensorBase:
 
         """
         tensors = [self._data, self._transpose]
-        self._data = None
-        self._transpose = None
         return tensors, self
 
     def restore_from_saved(

transformer_engine/pytorch/tensor/_internal/mxfp8_tensor_base.py

@@ -100,8 +100,6 @@ class MXFP8TensorBase:
 
         """
         tensors = [self._rowwise_data, self._columnwise_data]
-        self._rowwise_data = None
-        self._columnwise_data = None
         return tensors, self
 
     def restore_from_saved(