tally using the volumes instead of materials

[shimwell]

the use of materials to perform cell tallies means that unique materials are needed for each unique cell tally. This can be a little restrictive and can become complicated when the user wants to use the same material for separate volumes.

Perhaps we should move to tally on volumes. The volumes ids can be identified using mbgsets like this:

mbgsets dagm.h5m | grep 'Volume'

ps thanks to Andy for showing me that command.

We would also need a method of linking the materials to the original geometry.

With there pymoab route there is no imprinting and merging so I am hoping that volumes are just added in and counted in the same order.

With the Trelis route we already keep track of the volumes and save it in the geometry_details.json file

I shall take a look into this when I get a chance