Add support for Riemann boundary condition to all scalar solvers

[aureliegrebe]

Proposed Changes

Add support for Riemann boundary conditions to all scalar solvers by moving BC_Riemann definition from CTurbSolver to CScalarSolver.

I've only tested this for for CSpeciesSolver; I put the definition in CScalarSolver mainly just to avoid code duplication between CTurbSolver and CSpeciesSolver. Please let me know if there is a reason not to do this.

PR Checklist

• I am submitting my contribution to the develop branch.
• My contribution generates no new compiler warnings (try with --warnlevel=3 when using meson).
• My contribution is commented and consistent with SU2 style (https://su2code.github.io/docs_v7/Style-Guide/).
• I used the pre-commit hook to prevent dirty commits and used pre-commit run --all to format old commits.
• I have added a test case that demonstrates my contribution, if necessary.
• I have updated appropriate documentation (Tutorials, Docs Page, config_template.cpp), if necessary.

[pcarruscag]

Thanks for this — moving BC_Riemann up to CScalarSolver to share it across turbulence and species is a nice simplification and the change itself looks correct as a relocation (dispatch to BC_Inlet/BC_Outlet is virtual, so it resolves properly for each solver).

While reviewing, we ran into a pre-existing gap that your change surfaces more visibly now that species goes through this path too: scalar inlet values (turbulence/species) are never actually populated for a MARKER_RIEMANN boundary, so BC_Riemann silently falls back to freestream turbulence/species values at that marker regardless of what's imposed on the flow side.

The root cause is in the inlet-filling logic, which only populates the per-marker scalar arrays when the marker's KindBC is INLET_FLOW (or SUPERSONIC_INLET for species) — a RIEMANN_BOUNDARY marker never qualifies:

• CSpeciesSolver::SetUniformInlet — SU2_CFD/src/solvers/CSpeciesSolver.cpp:521
• CTurbSASolver::SetUniformInlet — SU2_CFD/src/solvers/CTurbSASolver.cpp:1954

For species this means there's currently no way to specify the composition entering through a Riemann boundary at all — it's silently freestream regardless of MARKER_INLET_SPECIES. Turbulence has the same issue (pre-existing, not introduced by you) since CConfig::GetInlet_TurbVal falls back to freestream when the tag isn't found under MARKER_INLET_TURBULENT, which it never will be for a Riemann-tagged marker.

Could you extend the fix to make these SetUniformInlet (and equivalent) checks also apply to RIEMANN_BOUNDARY markers, so MARKER_INLET_TURBULENT/MARKER_INLET_SPECIES entries are actually honored there?

Lastly, could you add a test case (or adapt an existing regression case) that uses MARKER_RIEMANN with the species/turbulence solver, so this path gets exercised in CI going forward?

[bigfooted]

@joshkellyjak can you have a look as well?

[aureliegrebe]

While reviewing, we ran into a pre-existing gap that your change surfaces more visibly now that species goes through this path too: scalar inlet values (turbulence/species) are never actually populated for a MARKER_RIEMANN boundary, so BC_Riemann silently falls back to freestream turbulence/species values at that marker regardless of what's imposed on the flow side.

I've added logic to the CSpeciesSolver::SetUniformInlet that sets the inlet values according to MARKER_INLET_SPECIES if the Riemann boundary is treated as an inlet.

For species this means there's currently no way to specify the composition entering through a Riemann boundary at all — it's silently freestream regardless of MARKER_INLET_SPECIES. Turbulence has the same issue (pre-existing, not introduced by you) since CConfig::GetInlet_TurbVal falls back to freestream when the tag isn't found under MARKER_INLET_TURBULENT, which it never will be for a Riemann-tagged marker.

I'm not entirely convinced there was actually a bug in the turbulent solvers causing MARKER_INLET_TURBULENT to be ignored. This value is never read in SetUniformInlet under any case, and is read in the BC_Inlet, which BC_Riemann calls. As far as I can tell, the value set in SetUniformInlet is only used when inlet-files are used, which is not supported for Riemann boundary conditions.

Lastly, could you add a test case (or adapt an existing regression case) that uses MARKER_RIEMANN with the species/turbulence solver, so this path gets exercised in CI going forward?

I've copied and modified the axi_rans_air_nozzle_restart to create axi_rans_air_nozzle_species.

[bigfooted]

@aureliegrebe Thank you for this contribution!