Add clients over time data structure

[DL6ER]

By submitting this pull request, I confirm the following (please check boxes, eg [X]) Failure to fill the template will close your PR:

Please submit all pull requests against the development branch. Failure to do so will delay or deny your request

• I have read and understood the contributors guide.
• I have checked that another pull request for this purpose does not exist.
• I have considered, and confirmed that this submission will be valuable to others.
• I accept that this submission may not be used, and the pull request closed at the will of the maintainer.
• I give this submission freely, and claim no ownership to its content.

How familiar are you with the codebase?:

10

---

See pi-hole/web#583

This template was created based on the work of udemy-dl.

[DL6ER]

Apply API_EXCLUDE_CLIENTS filtering as suggested by @WaLLy3K in private communication.

No filtering:

Assume I want to filter:

• localhost (bright blue) as it is constantly active
• desktop (darker blue starting around 20:00) just to demonstrate it works.

I apply the following filter:

where I intentionally used the host name for localhost and the IP address for desktop to show that both can be used at liberty to identify individual clients.

The result with the above filter is:

The contributions from localhost and desktop are now missing in the graph. Note that the colors between the two screenshots have changed as the chart generator isn't even aware of the existence of the two filtered clients.

This seems to be useful in some situations, e.g. when there is a very chatty device that likes to call home and hence totally ruins the overall statistics without filtering. In this situation, the new clients over time graph will be a good alternative to the main graph.

[AzureMarker]

The web interface should change its description from just top clients to include this feature.

[WaLLy3K]

Some of this is FTL's scope, other parts of this review are in scope for AdminLTE #583:

The popover gets cut off when there's more than a handful of clients, and there's an entry with a blank name - perhaps because there's two devices on my network which haven't had hostnames manually assigned to them via /etc/hosts

I also feel that for usability and readability's sake, that the devices should be sorted descending by the number of queries they've made. This makes it easier to hover over one data point, look at the top of the list and immediately say "Okay, the insane number of queries came from XYZ".