Normalization options

[CodyCBakerPhD]

Related to #9

For certain data views, such as 'by asset', normalize each by asset size (since some bias exists in the histogram relative to asset size and thus not really showing which assets are 'more popular' than others

For when 'number of requests' becomes a metric, also the 'number of objects' within the file could also be a normalization basis

[CodyCBakerPhD]

Though perhaps 'normalization' is the wrong word; perhaps 'scale by' and the basis itself is a selectable option?

[CodyCBakerPhD]

Similar to #7 requires a data source for the size or other measures associated with a Dandiset or asset