Insights Time Series Graph should drop to 0 instead of keeping last value for empty groups

[Markus-Brand]

Describe the Improvement

In the Insights -> Time Series Graph, only groups with data points in them create a new point in the graph.
Therefore, when there are no items in a group, the graph (for stepline curves) just stays at the level of the last group that had data. This is very misleading for many usecases and the value should drop to 0 for these groups.

Example:
We try to visualize occurence of certain events, but the same goes for any example with sparse data.
This graph contains 5 events: 3 at the beginning and then after a few minutes 2 more events 1 minute apart. The precision is set to minutes. This is what I get:

It looks like there are 3 events every minute until it drops to 1 event per minute. (and a graph with that data would be indistinguishable)
This is what I would expect this graph to look like:

This would represent the actual data way better.

[hanneskuettner]

Can you share an example collection setup for your datapoints as well as some seed data, so we can easily reproduce this?

[Markus-Brand]

Sure, this is the setup.

Seeding data is as simple as it gets: just creation_date and id.
seeding.csv