Add ranges to more performance test assertions

The last six performance test runs didn't reproduce this, only seeing it on full test runs:

Quick way to check is where it says the number of spans - i.e. '62 spans' on each of these.

https://grafana.prod.cluster.tag1.io/explore?panes=%7B%22d4B%22:%7B%22da...
https://grafana.prod.cluster.tag1.io/explore?panes=%7B%22hFh%22:%7B%22da...
https://grafana.prod.cluster.tag1.io/explore?panes=%7B%22b1m%22:%7B%22da...
https://grafana.prod.cluster.tag1.io/explore?panes=%7B%22AZk%22:%7B%22da...
https://grafana.prod.cluster.tag1.io/explore?panes=%7B%225GG%22:%7B%22da...

If it helps capture more info sure. Only one that I was going back n forth on was the range 66-67 but may just be lack of coffee at the moment. Don't see any breaking feature though.

Thinking now it might be worth postponing this on 📌 Allow needs_destruction services to run on page cache hits Needs review just in case that issue makes things more reliable overall.

Just in case...

Starting again here since that issue hasn't removed the variation but all the numbers have changed now we're recording differently. @andypost noticed one on https://git.drupalcode.org/issue/drupal-839444/-/jobs/725475

Not seeing any variation in the traces from https://grafana.prod.cluster.tag1.io/d/teMVIdjVz/umami?orgId=1&refresh=3...

It's possible this is infrequent so just isn't showing up in the last six traces, or it's possible that it only shows up on a full test run. If it's the latter, that will be very annoying to track down.

Used to run locally few times and it always passing

Committing this one from NR as it's a one digit change and I've seen it as a false positive a couple of times already.

Committed 110c3f0 and pushed to 11.x. Thanks!

longwave → closed merge request !6385

Automatically closed - issue fixed for 2 weeks with no activity.