STY: is_ci_environment seems like an antipattern

[jbrockmendel]

is_ci_environment is recent-ish and I trust it was implemented for a good reason, but it still isnt a good pattern long-term. This is a placeholder issue to track down the places where it is used and try to make it unnecessary. e.g. if a test uses more RAM than is available on the CI, a better check might be specifically for low-ram environments.

[mroeschke]

I believe I added it as a way to ensure tests weren't skipped locally just because it didn't work in the CI. You're correct that generally a more specific check could be used, but it may be difficult to make a run time check for some of the conditions (e.g. flakiness, services available)