"Fix" bogus idempotent operations warning due to loop unrolling not unrolling enough loops to show that an invariant
doesn't hold. This fix is to increase the loop unrolling count to 4, which experiments show doesn't typically impact
analysis time. The real fix is to modify the IdempotentOperationsChecker to suppress warnings where an analysis point
could be preceded by a point where we gave up due to loop unrolling.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@116769
91177308-0d34-0410-b5e6-
96231b3b80d8