Improve pipeline counts and visualisations

[apriltuesday]

Categorise each portion of ClinVar processed by the pipeline to provide visual explanation of the impact each part of the pipeline has on the output.

Some initial thoughts:

• better documentation or visualisation of criteria for ClinVar records being excluded - see #278
• maybe an expanded version of mapping statistics spreadsheet but more visual and more public
• consider impact of submission data on different attributes of ClinVar data, e.g. as presented in the variant type sankey diagrams, to get an idea of trends and inform decision making

[apriltuesday]

Another thing we've discussed regarding this is to incorporate the sankey diagram creation into the regular running of the pipeline. This can ensure that we run it regularly and also allows us to add to it, e.g. which variant types are processed by which consequence pipeline.