fix(post): exchange chemistry temperature seam ghosts for chemistry

[sbryngelson]

post_process converts conservative → primitive variables over the ghost-inclusive bounds, so the temperature Newton guess must be valid at rank seams for every chemistry run. Previously the temperature seam ghost was exchanged only when chem_params%diffusion was set, so a multi-rank non-diffusion chemistry run left the rank-seam Newton guess uninitialized → NaN temperature / pressure / sound-speed in the output.

Fix:

• m_mpi_common: exchange the temperature seam ghost for any chemistry run in post_process (simulation keeps the diffusion-only condition, since it does not convert over the ghost shell) via an #ifdef MFC_SIMULATION split.
• post_process/m_start_up: seed q_T over the interior (idwint) rather than the ghost-inclusive bounds (idwbuff) — the ghost q_cons is unread at that point, so a ghost-inclusive Newton sweep iterates on uninitialized data; s_populate_variables_buffers extends q_T into the ghosts afterward.

Extracted as a standalone fix from the AMR/scaling work (up/mega); the AMR-specific coupling in the original change was dropped for this general fix. CI runs the full matrix as the gate.

[Copilot]

Pull request overview

Fixes a post-processing MPI seam/ghost-cell initialization issue for chemistry runs by ensuring the temperature Newton initial guess (q_T) is valid at rank seams before conservative→primitive conversion is performed over ghost-inclusive bounds. This prevents uninitialized-temperature propagation that can yield NaN temperature/pressure/sound speed in multi-rank, non-diffusion chemistry post-processing runs.

Changes:

• Seed chemistry temperature (q_T_sf) over interior bounds (idwint) in post_process startup, then extend into ghosts via the existing buffer-population step.
• In m_mpi_common, exchange temperature seam ghosts for any chemistry run in non-simulation builds, while keeping the diffusion-only exchange behavior in simulation via an #ifdef MFC_SIMULATION split.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
src/post_process/m_start_up.fpp	Seeds q_T_sf over interior bounds to avoid Newton iteration on unread ghost q_cons, relying on later buffer population to fill ghosts.
src/common/m_mpi_common.fpp	Ensures q_T_sf seam ghost exchange occurs for all chemistry post-processing runs (not only diffusion), while preserving simulation’s diffusion-only behavior.

[codecov]

Codecov Report

❌ Patch coverage is 50.00000% with 1 line in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (master@aed0d74). Learn more about missing BASE report.
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
src/common/m_mpi_common.fpp	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff            @@
##             master    #1649   +/-   ##
=========================================
  Coverage          ?   59.56%           
=========================================
  Files             ?       83           
  Lines             ?    21112           
  Branches          ?     3128           
=========================================
  Hits              ?    12576           
  Misses            ?     6434           
  Partials          ?     2102           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.