fluids: Update PetscCallCeed (from Ratel)

Update from: https://gitlab.com/micromorph/ratel/-/merge_requests/815

Merge pull request #1562 from CEED/jrwrigh/fluids_problem_pointer

fluids: Make ProblemData a pointer to struct

fluids: Make ProblemData a pointer to struct

We basically use it as a pointer-to-struct type everywhere, so might as
well make it consistent with everything else

fluids - use VecGetState

Merge pull request #1558 from CEED/jrwrigh/smartsim_multiple_filter_widths

fluids: Add option for multiple filter widths during in-situ SGS training

fluids: moved SmartRedis send of time step number after all training data is computed

fluids: Remove ceed backend from advection.yaml

Due to ordering, this overides the regression tests

fix(fluids): Fix primary PETSc memory leak, others

All the memory leaks reported in regression tests via `-malloc_dump`
are now related to the boundary conditions and their handling of
QFunctionCont

fix(fluids): Fix primary PETSc memory leak, others

All the memory leaks reported in regression tests via `-malloc_dump`
are now related to the boundary conditions and their handling of
QFunctionContexts, which is a much more difficult problem and will be
addressed later.

show more ...

fluids: Rename PetscSmartRedisCall -> PetscCallSmartRedis

fluids: Compute DiffFilter RHS only if solution changes

fluids: Multiple filter widths for SGS training

fluids: Misc cleanup of SmartSim and SGS training

test(fluids): Test diffusion in advection

fluids: Add diffusion to advection problem

Merge pull request #1553 from CEED/jrwrigh/ns_stabilized_mass

fluids: Add stabilized mass operator for Newtonian

fix(fluids): Renable SmartSim testing and fix bug

- Fixed bug in sgs_dd_train QF, left over from debugging
- SmartSim tests were previously not running at all, due to
junit.py checking if `SMARTRE

fix(fluids): Renable SmartSim testing and fix bug

- Fixed bug in sgs_dd_train QF, left over from debugging
- SmartSim tests were previously not running at all, due to
junit.py checking if `SMARTREDIS_DIR` was a file instead of a
directory

Co-authored-by: James Wright <james@jameswright.xyz>

show more ...

Merge pull request #1533 from CEED/jrwrigh/smartsim_array_size

fluids: Change SmartSim sizeInfo to show minimum global array size

test(fluids): Test newtonian stabilized mass

refactor(fluids): Use QdataUnpack in Newtonian RHS

fluids: Add stabilized mass matrix to newtonian

Testing with:
```
build/fluids-navierstokes -options_file ./examples/fluids/gaussianwave.yaml -dm_plex_box_faces 10,10,1 -ts_max_steps 500 -degree 2

fluids: Add stabilized mass matrix to newtonian

Testing with:
```
build/fluids-navierstokes -options_file ./examples/fluids/gaussianwave.yaml -dm_plex_box_faces 10,10,1 -ts_max_steps 500 -degree 2 -implicit false -ts_type rk -stab supg -ts_dt 1e-4 -implicit false -ts_monitor_solution cgns:gaussian_wave.cgns -ts_monitor_solution_interval 5 -ts_view -state_var conservative -mass_ksp_type gmres -ts_monitor -mass_pc_type jacobi -mass_pc_jacobi_type diagonal
```

show more ...

Merge pull request #1554 from CEED/jrwrigh/petsc_fix_localization

fluids: Update topological periodicity localization

fluids: Update topological periodicity localization

See https://github.com/CEED/libCEED/issues/1470 for details.
Relies on PETSc MR in https://gitlab.com/petsc/petsc/-/merge_requests/7438

tl;dr is

fluids: Update topological periodicity localization

See https://github.com/CEED/libCEED/issues/1470 for details.
Relies on PETSc MR in https://gitlab.com/petsc/petsc/-/merge_requests/7438

tl;dr is that the localization of the continuous coordinate DM to the
discontinuous coordinate DM (i.e. the cell DM) must be done *after* the
coordinate DM is set to the tensor product ordering

show more ...

test(fluids): Add test for advection stabilized mass

ie. SUPG

fluids: Add stabilized mass for advection

This enables "full" SUPG for the advection problem using explicit time
integrators

fix(fluids): Allow advection to work with SYCL backends