Upgrade list_inactive_datasets()

[jeanpaulrsoucy]

Currently list_inactive_datasets() simply lists any dataset that has not been updated in at least 7 days. However, many datasets which have an update schedule longer than this (e.g., every 4 weeks) are listed in this output, even though they are not inactive datasets. Instead, the function should compare the the current run of duplicate datasets to the maximum length of previous runs of duplicate datasets and only report a dataset as potentially inactive if the current run exceeds the length of previous runs.

For example, if the current run of duplicates has lasted 8 days but the previous maximum length of a run was 28 days (e.g., because the dataset has a 4-week update cycle), it would not be flagged until the current run of duplicates reached 29 days.