Improve testing of simulated astronomical calendar invariants #7121
Description
#7075 removes the test for simulated, but your change also removes coverage for all other tests that set iters to values less than 2000, and I don't know if that hides potential issues. Please restore.
Originally posted by @robertbastian in #7012 (comment)