implement save raw tensor API

[Vikas-kum]

Save raw tensor api needs to be implemented. Use case is - user wants to save a tensor which is not part of model graph.

The implementation would look like below -

def save_raw_tensor(tname, tval):
  if self.writer is None:
    self._initialize_writers()
    self._write_raw_tensor_simple(scalar_name, scalar_val)

[Vikas-kum]

For time being there is workaround and users can try this function in script code -

import smdebug.tensorflow as smd ## Please replace with mxnet/xgboost/torch according to framework you are using 
def save_raw_tensor(hook, tname, tval):
  if hook.writer is None:
    hook._initialize_writers()
    hook._write_raw_tensor_simple(tname,  tval) 
hook = smd.get_hook(create_if_not_exists=True) 
save_raw_tensor(hook, tname, tval)
-----
-----
----
--
hook.close() # this is final statement, tells hook to write the step