insights: how to represent language stats time scope

I guess my only concern here is that the semantic of a 0 value interval is ambiguous. What does that mean to a client, and how should one execute the search queries to generate this series based on this information?

I wonder if we just should have a new TimeScope type for a single time point (the most recent for example)? The intention behind the TimeScope was to ultimately have multiple types, some of which would have an interval and others would be different. Maybe this is a justification here? What do you think?

Originally posted by @coury-clark in https://github.com/sourcegraph/sourcegraph/pull/27102#discussion_r743760401