Clean up function to fill missing data between beam datasets

[DavidMcCann-NOC]

Currently the function seastar.utils.tools.fill_missing_variables works well but its a bit hard to read, and there is probably a much cleaner, more pythonic way of achieving the same result.

A good idea would to be to create a var_list between all three beams in a triplet and then create NaN variables in beams that do not share the same variables.

[DavidMcCann-NOC]

This now exists in a cleaner way as of #224 , however still using fill_missing_variables

[DavidMcCann-NOC]

Closed with #231