amdflang: dist_schedule(static, 4) is not applied: 64 iterations over 4 teams go to teams 0 and 1 in two blocks of 32

! # Draft finding: omp52-clause-dist_schedule-chunk_size on afar
!
! - compiler: afar
! - toolchain: afar; arch: gfx942; machine: tuolumne; memory mode: unified; opt: -O2
! - program: catalog omp52-clause-dist_schedule-chunk_size (f90, version 52, sha256 3628c7a53985f9c2)
! - expected: check=pass (trusted; derivation + model + probe)
! - 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 (derivation + model + probe): stored main verdict: derivation + model + probe agree;
!   witnesses: afar/gfx942/copy (port:omp52-c-clause-dist_schedule-chunk_size), afar/gfx942/unified
!   (port:omp52-c-clause-dist_schedule-chunk_size), cce19/gfx942/copy, cce19/gfx942/unified, cce20/gfx942/copy,
!   cce20/gfx942/unified, cce21/gfx942/copy, cce21/gfx942/unified
! - 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-dist_schedule-chunk_size), afar/gfx942/unified
!   (port:omp52-c-clause-dist_schedule-chunk_size), cce19/gfx942/copy, cce19/gfx942/unified, cce20/gfx942/copy,
!   cce20/gfx942/unified, cce21/gfx942/copy, cce21/gfx942/unified
! - withdrawn: none
!
! Generated by compliance.gen (OpenMP 5.2, free-form Fortran). Do not edit.
! id: omp52-clause-dist_schedule-chunk_size
! primary atom: omp.clause.dist_schedule.chunk_size
! legal: yes
! observable: yes
! ran_on_host: inside the target region of interest
! evidence: construct
! needs: none
! std: f2008
! derivation (check=pass): dist_schedule(static, 4): chunks of 4 iterations assigned to the initial teams round-robin in team-number
! derivation: order (5.2 §11.6.1 p.256), so iteration i runs on team mod((i - 1)/4, nt), which dok tests; nt >= 1 and b = ref; the
! derivation: check accepts
program gen_prog
  use omp_lib, only: omp_get_num_teams, omp_get_team_num, 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 :: team(n), nt
  logical :: dok
  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
  team = -1
  nt = -1
  !$omp target teams distribute parallel do map(to: a) map(from: b) map(from: team, nt) &
  !$omp& map(tofrom: host_flag) default(none) shared(a, b, team, nt, host_flag) num_teams(4) &
  !$omp& dist_schedule(static, 4)
    do i = 1, n
      b(i) = 2*a(i) + 1
      team(i) = omp_get_team_num()
      if (i == 1) nt = omp_get_num_teams()
      if (i == 1) host_flag = merge(1, 0, omp_is_initial_device())
    end do
  !$omp end target teams distribute parallel do
  dok = .true.
  do i = 1, n
    dok = dok .and. team(i) == mod((i - 1)/4, nt)
  end do
  ok = all(b == ref) .and. (nt >= 1) .and. dok
  print '(a,i0,a,a)', 'RESULT ran_on_host=', host_flag, ' check=', merge('pass', 'fail', ok)
  deallocate(a, b)
end program gen_prog