Tidy up the ROPP netCDF global attributes

[Ian Culverwell]

ROPP8.0 beta reviewer Axel von Engeln (EUM) says:

  - My first runs that did not set the attribute institution seem 
to result in the bg_source being put there. Is that possible?
  Response: I don't think so. When I remove global attribute 
"institution" from test file and try again, it just fails with:

I think I didn't really wrote this too clearly. I was merely not 
setting the institute in the global attributes section. I was 
empty. And I just verified that again and it seems there is still a 
problem (although different now). When I run the attached bg file 
to a fm file, the institute is not kept, but just empty in the fw 
file. Here is my command line: ropp_fm_bg2ro_1d 
GRAS_Institution_Test.bg.nc -o 
GRAS_Institution_Test.fw.nc. The bg file is attached. Do you get 
the same?

I also noted that some generic text is put into the title of the 
global attributes for the generated fw file. Maybe that can be done 
better by e.g. pointing to the tool that was used to generate the 
file.

I confirm I get the same result, because:

1. ROPP's output global attribute 'institution' = the input 'processing_centre', or 'bg_source' if you're outputting background data. The input 'institution' is ignored. Why? This doesn’t make sense to me. Surely the output should mirror the input as closely as possible.
2. The global attribute 'title', is hardwired to 'ROPP Radio Occultation data' or 'Atmospheric background data for ROPP Radio Occultation data' if you're outputting background data. Again, the 'title' in the input file is just ignored. Christ.

The whole set of ROPP global attributes needs examining. Note the overlap with #382.

[Ian Culverwell]

GRAS_Institution_Test.bg.nc

[Ian Culverwell]

Defer to ROPP10.0.

[Ian Culverwell]

Some work by DMI on this. Defer to ROPP-11.0.

[Ian Culverwell]

Rolling over to ROPP-12.0.