Damp hyper-cleaning psi with the current-stage value #2064

The psi damping term in s_compute_rhs read q_prim_vf(psi), the time
stepper's primitive array. That array is only refreshed at the end of
s_compute_rhs, and only when run_time_info, probe_wrt, ib or
bubbles_lagrange is set, so the damping used the previous stage's psi
or, with none of those flags, a never-written array. Toggling
run_time_info changed the solution.

Use q_prim_qp%vf(psi), which aliases the current-stage conservative psi.
run_time_info = T and F now give bit-identical output on a shrunk
2D_orszag_tang_hyper_cleaning case.

Regenerated golden for B4DC99F9 (2D -> MHD -> hyper_cleaning); max
abs deviation 3.9e-4.

A fix for this is already written. The pull request was closed only to reduce the number of simultaneously open PRs from this author, not because the work was rejected or found incorrect.

• Closed PR: Damp hyper-cleaning psi with the current-stage value #1987
• Branch: fix/hyper-cleaning-psi on the sbryngelson fork, at 6365acbad4
• Reopening needs no force-push and no rebase: gh pr reopen 1987 -R MFlowCode/MFC

Tracked with the pr-to-do label so the whole set is discoverable together.

Filed with Claude Code.