"GRIB_PARAM" attributes are not saving to netcdf

[pp-mo]

User reported locally that this appeared to have changed with new versions of iris-grib,
and I originally thought was due to extension of the "GRIB_PARAM" attribute scheme to GRIB2 data in the latest v0.20 release.

But actually it doesn't work with GRIB2 data either,
so I now think this is actually a bug in newer versions of Iris, i.e. (probably) v3.11
-- TBC

Example:

>>> import iris
>>> from iris.tests import get_data_path
>>> 
>>> pth2 = itst.get_data_path(["GRIB", "time_processed", "time_bound.grib2"])
>>> 
>>> cube2 = iris.load_cube(pth2)
>>> print(cube2)
air_temperature / (K)               (latitude: 73; longitude: 96)
    Dimension coordinates:
        latitude                             x              -
        longitude                            -              x
    Scalar coordinates:
        forecast_period             -11055.0 hours, bound=(-28587.0, 6477.0) hours
        forecast_reference_time     1998-03-06 03:00:00
        pressure                    100000.0 Pa
        time                        1996-11-30 12:00:00, bound=(1994-12-01 00:00:00, 1998-12-01 00:00:00)
    Cell methods:
        0                           time: mean
    Attributes:
        GRIB_PARAM                  GRIB2:d000c000n000
>>> 
>>> iris.save(cube2, "tmp2.nc")
/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py:2668: IrisDeprecation: Saving to netcdf with legacy-style attribute handling for backwards compatibility.
This mode is deprecated since Iris 3.8, and will eventually be removed.
Please consider enabling the new split-attributes handling mode, by setting 'iris.FUTURE.save_split_attrs = True'.
  warn_deprecated(message)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/io/__init__.py", line 477, in save
    result = saver(source, target, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 2750, in save
    sman.write(
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 647, in write
    self.update_global_attributes(global_attributes)
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 712, in update_global_attributes
    _setncattr(self._dataset, attr_name, attributes[attr_name])
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 271, in _setncattr
    return variable.setncattr(name, attribute)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "src/netCDF4/_netCDF4.pyx", line 3087, in netCDF4._netCDF4.Dataset.setncattr
  File "src/netCDF4/_netCDF4.pyx", line 1888, in netCDF4._netCDF4._set_att
TypeError: illegal data type for attribute b'GRIB_PARAM', must be one of dict_keys(['S1', 'i1', 'u1', 'i2', 'u2', 'i4', 'u4', 'i8', 'u8', 'f4', 'f8']), got O
>>> 

[pp-mo]

Actually I think no arbitrary objects save to netcdf.

>>> import iris
>>> print(iris.__file__)
/home/h05/itpp/git/iris/iris_main/lib/iris/__init__.py
>>> print(iris.__version__)
3.11.1
>>> if hasattr(iris.FUTURE, "save_split_attrs"):
...     iris.FUTURE.save_split_attrs = True
... 
>>> from iris.cube import Cube
>>> class MyObj:
...   x: int = 3
... 
>>> cube = Cube(
...   [1],
...   attributes={
...     'x': MyObj()
...   }
... )
>>> iris.save(cube, "tmp.nc")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/io/__init__.py", line 477, in save
    result = saver(source, target, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 2750, in save
    sman.write(
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 589, in write
    cf_var_cube = self._create_cf_data_variable(
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 2249, in _create_cf_data_variable
    _setncattr(cf_var, attr_name, value)
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/saver.py", line 271, in _setncattr
    return variable.setncattr(name, attribute)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/h05/itpp/git/iris/iris_main/lib/iris/fileformats/netcdf/_thread_safe_nc.py", line 122, in setncattr
    return self._contained_instance.setncattr(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "src/netCDF4/_netCDF4.pyx", line 4604, in netCDF4._netCDF4.Variable.setncattr
  File "src/netCDF4/_netCDF4.pyx", line 1762, in netCDF4._netCDF4._set_att
TypeError: illegal data type for attribute b'x', must be one of dict_keys(['S1', 'i1', 'u1', 'i2', 'u2', 'i4', 'u4', 'i8', 'u8', 'f4', 'f8']), got O
>>>   

[pp-mo]

There is special support for STASH objects : here
But not for anything else