Generalise to GTMD Distributions

[Radonirinaunimi]

The following PR completes the non-perturbative distributions by adding supports to GTMDs (and by consequence GPDs). To ensure backward compatibility, it depends on neopdf@0.2.0 for the parsing of the v1 grids.

TODO:

• account for $(\xi, \Delta)$ inputs when writing new grids
• perform extensive benchmark
• remove version specification in benchmark CI
• test Python 3.14 on the CI
• Extend Chebyshev interpolations to 4D and 5D

[github-actions]

🚀 Benchmark Results 🚀

Here are the results from the benchmark run.

xfxq2                   time:   [70.152 ns 70.213 ns 70.299 ns]
Found 9 outliers among 100 measurements (9.00%)
  1 (1.00%) low mild
  4 (4.00%) high mild
  4 (4.00%) high severe

xfxq2s                  time:   [7.2462 µs 7.3974 µs 7.5973 µs]
Found 15 outliers among 100 measurements (15.00%)
  3 (3.00%) high mild
  12 (12.00%) high severe

xfxq2_members           time:   [621.19 ps 621.26 ps 621.34 ps]
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) high mild
  3 (3.00%) high severe

xfxq2_cheby             time:   [358.71 ns 359.21 ns 359.93 ns]
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) high mild
  5 (5.00%) high severe

xfxq2_cheby_batch       time:   [1.9852 µs 1.9914 µs 1.9981 µs]
Found 11 outliers among 100 measurements (11.00%)
  3 (3.00%) high mild
  8 (8.00%) high severe

[github-actions]

🚀 Benchmark Results 🚀

Here are the results from the benchmark run.

xfxq2                   time:   [67.222 ns 67.299 ns 67.399 ns]
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) high mild
  3 (3.00%) high severe

xfxq2s                  time:   [6.9279 µs 6.9399 µs 6.9561 µs]
Found 5 outliers among 100 measurements (5.00%)
  1 (1.00%) high mild
  4 (4.00%) high severe

xfxq2_members           time:   [622.30 ps 623.30 ps 624.55 ps]
Found 19 outliers among 100 measurements (19.00%)
  2 (2.00%) high mild
  17 (17.00%) high severe

xfxq2_cheby             time:   [359.20 ns 359.56 ns 360.02 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high severe

xfxq2_cheby_batch       time:   [1.9067 µs 1.9075 µs 1.9084 µs]
Found 20 outliers among 100 measurements (20.00%)
  3 (3.00%) low severe
  9 (9.00%) low mild
  5 (5.00%) high mild
  3 (3.00%) high severe

[Radonirinaunimi]

🔴 For the record, below are the benchmark results when comparing to the published v0.2.0 release. The lines represent successively the results for: LHAPDF and its std error from 20 independent runs, NeoPDF and its errors, and NeoPDF lazy loading and its errors.

v0.2.0

1.22412658e-01 1.48847425e-01 1.79906476e-01 2.16379547e-01 2.53767252e-01 2.94069660e-01 3.45752573e-01 3.84723699e-01 4.34436393e-01 4.89436328e-01 5.48730767e-01 6.12894654e-01 6.71056783e-01 7.32639420e-01 8.06839120e-01 8.89721894e-01 9.55257416e-01 1.04066429e+00 1.12291017e+00 1.21353761e+00
5.91819711e-03 2.05359698e-03 6.74703206e-03 6.13469358e-03 9.52976941e-03 6.03407658e-03 2.17744695e-02 5.32699161e-03 7.99714272e-03 7.36533110e-03 1.62440000e-02 2.62123149e-02 1.85770153e-02 1.00926159e-02 2.36521690e-02 4.18321102e-02 1.75208639e-02 2.81730530e-02 4.87574946e-02 4.55277668e-02
2.27455974e-02 2.85081983e-02 3.41438055e-02 4.11523581e-02 4.80696559e-02 5.61252117e-02 6.47661924e-02 7.42105246e-02 8.22400331e-02 9.25506949e-02 1.04438436e-01 1.16683900e-01 1.26823902e-01 1.40516138e-01 1.53568840e-01 1.68932128e-01 1.81399608e-01 1.98159981e-01 2.09479487e-01 2.27918482e-01
1.08409808e-03 1.08327278e-03 1.47185727e-03 2.63402705e-03 1.74889511e-03 1.72490104e-03 2.16508742e-03 2.71738090e-03 1.41341305e-03 1.30355056e-03 4.26102936e-03 5.38553011e-03 2.40785481e-03 5.66968541e-03 6.40362259e-03 9.54734895e-03 5.02875413e-03 8.19747227e-03 6.34622916e-03 8.90230329e-03
2.29035139e-02 2.88414001e-02 3.39885950e-02 4.14224625e-02 4.79324102e-02 5.58046699e-02 6.49387360e-02 7.37951040e-02 8.20215344e-02 9.29939508e-02 1.03920078e-01 1.17217076e-01 1.28154147e-01 1.40011108e-01 1.53805447e-01 1.68432164e-01 1.80347013e-01 1.98138154e-01 2.10349512e-01 2.28717923e-01
1.06790306e-03 2.49483021e-03 1.17939741e-03 2.21732522e-03 1.39934183e-03 9.40654336e-04 2.01186972e-03 2.21713371e-03 1.68461511e-03 2.73742810e-03 3.91461687e-03 5.68275553e-03 7.28565226e-03 4.12549214e-03 6.23252855e-03 8.00818368e-03 5.89788155e-03 8.47353447e-03 6.47613490e-03 7.49529552e-03

This branch

1.28283429e-01 1.57513583e-01 1.92902291e-01 2.28009260e-01 2.68884981e-01 3.11628139e-01 3.59393060e-01 4.07884753e-01 4.61679459e-01 5.11878431e-01 5.82192957e-01 6.45970798e-01 7.26056695e-01 7.99928284e-01 8.57821679e-01 9.14630508e-01 1.00384942e+00 1.09247993e+00 1.16982024e+00 1.24358894e+00
5.44152361e-03 6.74491732e-03 1.01778035e-02 9.49241297e-03 1.41591208e-02 1.66791649e-02 1.95327731e-02 1.86732322e-02 1.71098345e-02 6.49831742e-03 2.92235024e-02 4.39429136e-02 6.96656659e-02 9.71882464e-02 4.90904823e-02 1.95561690e-02 3.36248410e-02 4.11634350e-02 2.63567188e-02 1.61809201e-02
2.13468194e-02 2.64604211e-02 3.22004318e-02 3.80028844e-02 4.50765491e-02 5.26622415e-02 6.01440310e-02 6.79919839e-02 7.75335431e-02 8.63683105e-02 9.78528619e-02 1.10294867e-01 1.23727870e-01 1.33425331e-01 1.42742336e-01 1.54878700e-01 1.69998980e-01 1.82758617e-01 1.97740829e-01 2.13585079e-01
8.42706323e-04 1.58413361e-03 1.44304717e-03 1.38718911e-03 2.46752541e-03 3.02907167e-03 2.84061515e-03 1.26780253e-03 3.89037382e-03 2.34524702e-03 6.78656636e-03 1.09235949e-02 1.32157031e-02 1.20914344e-02 5.48475219e-03 3.76191395e-03 8.80737377e-03 7.59369072e-03 7.88376682e-03 9.65555654e-03
2.09879518e-02 2.64562607e-02 3.19406509e-02 3.78172159e-02 4.50649261e-02 5.19535422e-02 5.95540881e-02 6.77685022e-02 7.70904779e-02 8.69264483e-02 9.75015879e-02 1.10246718e-01 1.22964239e-01 1.32735217e-01 1.42034745e-01 1.52555692e-01 1.67977059e-01 1.80411184e-01 1.96289098e-01 2.12222850e-01
8.02489742e-04 1.40882081e-03 1.52288284e-03 1.14358651e-03 2.24088075e-03 2.27890881e-03 2.78761801e-03 1.33894525e-03 3.84508596e-03 4.19677102e-03 6.46599829e-03 1.21039125e-02 1.45139099e-02 1.17542740e-02 5.16840525e-03 1.80470772e-03 6.28256366e-03 5.80048345e-03 6.01290580e-03 9.09314284e-03