amdflang: ordered regions in a parallel do in a target region do not run in iteration order

! # Draft finding: omp52-clause-ordered on afar
!
! - compiler: afar
! - toolchain: afar; arch: gfx942; machine: tuolumne; memory mode: unified; opt: -O2
! - program: catalog omp52-clause-ordered (f90, version 52, sha256 50b2a05bbea64844)
! - expected: check=pass (trusted; elision + model)
! - actual: wrong_answer check=fail (check_fail: ran_on_host=0 check=fail)
! - verdict: compiler_bug by rule R20-runtime-bug: a legal program fails at run time, deterministically, against an expected value
!   from two or more independent sources with an executed witness, and another toolchain, mode, arch or the elision gives the
!   expected value
! - reason: expected trusted check=pass (elision + model): stored main verdict: elision + model agree; witnesses: afar/gfx942/copy
!   (port:omp52-c-clause-ordered), afar/gfx942/unified (port:omp52-c-clause-ordered), elision
! - citations: RUL-DERIVED-NEEDS-EXECUTION (rulings.md); RUL-WITNESS-PER-MODE (rulings.md);
!   RUL-SAME-COMPILER-RUNS (rulings.md)
! - legality: legal (legal by construction: offload.engine.gen / c_gen / acc_gen check every program against the catalog rules
!   and the
!   offload scope as it is built (no compiler))
! - witnesses: afar/gfx942/copy (port:omp52-c-clause-ordered), afar/gfx942/unified (port:omp52-c-clause-ordered), elision
! - withdrawn: none
!
! Generated by compliance.gen (OpenMP 5.2, free-form Fortran). Do not edit.
! id: omp52-clause-ordered
! primary atom: omp.clause.ordered
! legal: yes
! observable: yes
! ran_on_host: inside the target region of interest
! evidence: construct
! needs: none
! std: f2008
program gen_prog
  use omp_lib, only: omp_is_initial_device
  implicit none
  integer, parameter :: n = 64
  integer, allocatable, target :: a(:), b(:)
  integer :: ref(n)
  integer :: i, host_flag
  logical :: ok
  integer :: cnt, seq(n)
  allocate(a(n), b(n))
  do i = 1, n
    a(i) = i
    b(i) = 0
    ref(i) = 2*i + 1
  end do
  host_flag = 1
  cnt = 0
  seq = -1
  !$omp target map(to: a) map(from: b) map(tofrom: cnt) map(from: seq) map(tofrom: host_flag)
    host_flag = merge(1, 0, omp_is_initial_device())
    !$omp parallel do default(none) shared(a, b, cnt, seq) ordered
      do i = 1, n
        b(i) = 2*a(i) + 1
        !$omp ordered
          cnt = cnt + 1
          seq(i) = cnt
        !$omp end ordered
      end do
    !$omp end parallel do
  !$omp end target
  ok = all(b == ref) .and. all(2*seq + 1 == ref)
  print '(a,i0,a,a)', 'RESULT ran_on_host=', host_flag, ' check=', merge('pass', 'fail', ok)
  deallocate(a, b)
end program gen_prog
! spec (requirement): Ordered regions of a worksharing loop with an ordered clause execute sequentially in the order
!   of the loop iterations.
! "the threads in the team that is executing the worksharing-loop region execute ordered regions sequentially in the
!   order of the loop iterations." (5.2 §15.10.2 p.330 l.21-22)
! spec (requirement): An ordered construct without clauses has the effect of the threads clause.
! "If no clauses are specified, the effect is as if the threads parallelization-level clause was specified." (5.2
!   §15.10.2 p.330 l.20-21)
! spec (this program): The ordered region increments cnt and records seq(i), so the regions run in iteration order and
!   seq(i) = i for every i; on the device amdflang runs them out of order and check=fail results.