#403 removed the complexity of allowing record_failed invocations to override the participants specified when initiating the check. However, the method I used to try and find all the existing record_failed invocations that specified participants= was not effective as I missed many places. This has caused a failure in CI for an open PR, making it more difficult to identify the root cause of that failure. This PR attempts to clean up the remaining locations.

I identified the locations in this PR by searching all files for instances of participants= and manually evaluating each instance before removing it. I attempted to remove an instance of participants= specification only if 1) it was part of a record_failed invocation and 2) this argument was redundant with/identical to the participants specified when the check was initiated. (side note: we had a lot of redundancy!) I did not find any instances of record_failed where participants was overridden with a different or new value from when the check was initiated.

I also searched for instances of requirements= that may have been missed in a similar way, but I did not find any.

I did not perform a manual search of the 264 instances of record_failed we have in the codebase to identify any invocations that might have passed participants positionally rather than by keyword, but I intend to keep an eye out when we remove specification of severity from these invocations per #404.