export ql monoenergetic d11 and enable hdf5 efinal output

[krystophny]

Summary

Export the tokamak monoenergetic D11_ov_Dpl quantity from NEO-2-QL/efinal.h5 and expose prop_fileformat in the QL propagator namelist so HDF5 efinal.h5 can be requested from normal QL runs.

Why

The AUG 30835 GORILLA_APPLETS vs NEO-2 axisymmetric comparison needs a clean monoenergetic QL output contract. Before this change:

• NEO-2-QL computed D11_ov_Dpl-relevant quantities internally but did not write D11_ov_Dpl to efinal.h5
• prop_fileformat existed in propagator_mod but was not readable from NEO-2-QL/neo2.f90, so requesting HDF5 efinal.h5 via the input file failed

Changes

• add prop_fileformat to the QL /propagator/ namelist in NEO-2-QL/neo2.f90
• compute D11_ov_Dpl in NEO-2-QL/propagator.f90 using the same local pattern already used on the PAR side
• write D11_ov_Dpl into efinal.h5

Verification

$ cmake --build /home/ert/code/NEO-2/build -j$(nproc) --target neo_2_ql.x
[12/12] Linking Fortran executable NEO-2-QL/neo_2_ql.x

$ ./run_neo2.sh > run.log 2>&1
$ python /home/ert/data/AUG/NEO-2/30835/gorilla_axisymmetric_baseline/scripts/extract_neo2_par_monoenergetic.py \
    /home/ert/data/AUG/NEO-2/30835/gorilla_axisymmetric_baseline/runs/ql_two_surfaces/es_0p25271/efinal.h5
{
  "dmono_over_dplateau": -3.313532670952886e-05,
  "D11_ov_Dpl": -0.00018744171370511668
}

[krystophny]

Post-PR verification on the AUG 30835 axisymmetric benchmark harness is now in place.

Using the patched NEO-2-QL build, both reduced-surface runs completed and wrote efinal.h5 with D11_ov_Dpl:

• boozer_s = 0.2527070707070707: dmono_over_dplateau = -3.313532670952886e-05, D11_ov_Dpl = -1.8744171370511668e-04
• boozer_s = 0.4984141414141414: dmono_over_dplateau = -2.6360065981344247e-05, D11_ov_Dpl = -1.4911505126346674e-04

The same runs also provided the ion MtOvR entries used to match GORILLA_APPLETS v_E on the data side:

• boozer_s = 0.2527070707070707: MtOvR_ion = -6.838114130517892e-04
• boozer_s = 0.4984141414141414: MtOvR_ion = -1.6956511222229625e-03

Both matched GORILLA_APPLETS reruns then finished with zero lost particles.

The synchronized benchmark state is pushed in the data repo on branch ert/aug30835-gorilla-neo2-axisymmetric at commit 89d87aca.

[GeorgGrassler]

@krystophny quick question, why could you not use the neo2_multispecies_out.h5 to get your D11_ov_Dpl? This quantity is already available as D11_NA_Dpl there, assuming you do multispecies runs. Or is it specifically, because you wanna get the monoenergetic one? Cause you can also do that by just switching the settings of neo2.in to only use the Lorenz collision model.

[krystophny]

@krystophny quick question, why could you not use the neo2_multispecies_out.h5 to get your D11_ov_Dpl? This quantity is already available as D11_NA_Dpl there, assuming you do multispecies runs. Or is it specifically, because you wanna get the monoenergetic one? Cause you can also do that by just switching the settings of neo2.in to only use the Lorenz collision model.

true, just slopping around sry