Fix periodic axis geometry, slip normals and moving donor volumes

[rois1995]

Proposed Changes

Axis points kept an incomplete control volume, slip walls used a one-sided periodic normal, and mesh updates left donor volumes stale. This PR completes that periodic geometry.

Dependency: #2961. Its commits are included until that PR merges; the focused diff is rotation...axis.

Axis points now accumulate all rotated copies of the sector instead of keeping an incomplete control volume. Slip-wall normals merge raw face areas across periodic matches before normalization, including fine and coarse Euler walls. Mesh updates refresh donor volumes before later gradients and residuals use them.

The periodic3d_axis regression and partitioned unit checks cover wall normals and repeated deformation/Green–Gauss gradients. Its tiny mesh is in TestCases PR #210 on the branch fix_periodic_axis. The pre-existing workflow commit temporarily selects that branch and must be reverted before merging.

Moving-flow ALE/GCL and complete geometry sensitivities remain to be checked. Nonlinear dissipation on split faces can still differ from a full mesh. Complete regression-reference checks remain pending.

Validation of the combined periodic source passes serial, partitioned MPI2, OpenMP2 and MPI2×OpenMP2 (10 cases / 3175 serial assertions). Individual branch CI and complete regression/reference checks are pending. Test configurations, meshes, logs and before/after values are in the testcase comment. New regression references are local x86 values and need CI confirmation.

Related Work

• MPI Point-to-Point Refactoring + New Periodic BC Implementation #652 introduced the current periodic implementation.
• Fixes for rotational periodic boundaries: implicit solver, multigrid and limiters #2961 lists the eight related periodic-boundary PRs. Their numbering groups the topics; it does not imply a linear dependency chain.

PR Checklist

• I am submitting my contribution to the develop branch.
• My contribution generates no new compiler warnings (complete branch CI pending).
• My contribution is commented and consistent with SU2 style.
• I ran the repository pre-commit checks on the changed files.
• I have added tests that demonstrate the contribution.
• I have updated appropriate documentation, if necessary.

[rois1995]

Test cases

Files (configurations, histories, meshes, statistics, scripts): periodicBoundaries/axis. 2 MPI ranks, "develop" = 6db10127d1. The values of the new test are from my machine and will be updated from the CI.

Laminar pipe flow on a 30 degree sector with one cell around the axis, against the full pipe (no periodic boundary).

	axial velocity on the axis [m/s]	iterations to rms[Rho] = -12
sector, develop	69.44 at all axis points (the initial value)	4024
sector, this PR	35.20 to 39.28	2186
full pipe	35.24 to 39.28	1378

One iteration from the same field, largest difference to the full pipe at the same points (30 degree sector with one cell and 90 degree sector with three cells):

	develop	this PR
residuals, central scheme without dissipation	100 % at the axis points (their residual is removed), about 15 % at their neighbors	round-off at all points (1e-14 of the largest residual)
limiters, Green-Gauss	-	1e-12 at all points
limiters, least squares, 90 degree sector	-	1e-15 at all points

The same holds for a 180 degree sector (two copies).

With an upwind scheme or with JST the residual of periodic points differs from the full pipe also away from the axis (a few percent), because the flux over a face cut by the periodic boundary is computed in two parts. This is not changed here.

Convergence on the sector that is one cell wide (CFL 10): first-order ROE converges as in the table. Second order without limiter reaches rms[Rho] = -5.4 in 3000 iterations and needs CFL 2 to converge; JST stalls at -3.2 at CFL 10. The 90 degree sector with three cells converges with both (1538 and 2147 iterations, full pipe 1237 and 1314).

New test periodic3d_axis (first-order ROE, 100 iterations): rms[Rho] -3.290134 on develop, -3.353165 with this PR.

No existing reference value changes.

Additional tests (Oct 7)

Reproducers, configurations, numeric logs and scripts: geometry. Develop = 6db10127d1; BOX fixtures generate their meshes. The original B–D evidence above was recorded before the expanded follow-ups.

Check	develop	fixed combined source
Fine/coarse Euler slip-wall normal	2/4 assertions fail; fine error 0.065403 (3.75 degrees)	both levels pass
Repeated deformation donor volume	stored 0.0078125, current 0.0092605886	current donor volume after each update
Green–Gauss after deformation	max error 0.084817	passes linear-field check

These checks extend the original axis evidence above. ALE/GCL and complete moving-geometry sensitivities remain open.

Combined release checks: serial and OpenMP2 pass 10 cases / 3175 assertions; partitioned MPI2 and MPI2×OpenMP2 pass on both ranks (2334 / 2238 assertions). These are combined-source checks, not standalone builds of every branch. Complete branch CI and full regression/reference checks remain pending.