Decide how to handle (or not) duplicate exon entries

[standage]

For example, in case of alternative splicing, exons shared between isoforms can either be 1) duplicated or 2) have multiple Parent values.

[standage]

If we handle this, it should be a generator that consumes and yields entries of all types, but handles exon features (maybe arbitrary features?) correctly.

[standage]

For each feature CC create two dictionaries:

• feature by ID
• exon by range (defaultdict with range objects as keys and lists of exon entries as values)

For duplicated exons (lists with length > 1), add all parent IDs and add as child to appropriate parents.