Resolves#68096 , removing unnecessary scripted metric query from overview page and unifying the Check and Ping types.
This simplifies the types in a number of ways, and reduces the total quantity of code to execute the queries for the overview page. It also fixes the Tls and related types which were inconsistent and presented a problem here since they are used by this JS. There are now three stages where before there were four:
Find potential matches: where we determine which monitor IDs are eligible for the overview page
Refine potential matches: where we determine which ones actually match and return the summary documents for each location to build the MonitorSummary object
Get monitor histograms: where we calculate the histograms for each monitor. In the future we might make this a separate API call.
This improves the overall code structure, and leaves the test coverage about the same depending on how you look at it. I think we can do more to improve the quality of code / tests here, but this seemed like a good initial place to draw the line for now.
In perfunctory testing on our internal observability clusters I saw perf improve from 2.5s to 1.1s on the Uptime homepage with no filters. So, it looks like there are potentially perf improvements (no real benchmarking was done).
Previously, this returned all pings from the latest check group. This was not actually used anywhere, only the summary pings are required for the current UI, so we now only return those from the list API as this saves a query.
5085b62c9c