fix: ASE_POT compilation errors and lazy Python interpreter init

[HaoZeke]

Closes #324.

Four bugs in client/potentials/ASE/ASE.cpp:

• Constructor named ASEOrcaPot instead of ASE (copy-paste from ASE_ORCA)
• PotType::ASE_ORCA instead of PotType::ASE_POT
• isFPEEnabled/disableFPE/enableFPE don't exist; use eonc::FPEHandler
• Undeclared variable N; should be nAtoms

Additional fixes:

• Rename -DASE_POT to -DWITH_ASE_POT (macro clashed with PotType enum)
• Factory used make_shared<ASE_POT> instead of make_shared
• Lazy singleton interpreter (PyGuard.h) so non-ASE runs pay zero Python overhead; replaces per-site scoped_interpreter guards in ClientEON.cpp, GPSurrogateJob.h, and testpot.cpp
• ASE_ORCA and ASE_NWCHEM constructors call ensure_interpreter() for consistency
• Fix numpy include_directories() error on meson <1.10 by using compile_args instead
• Add ASE LJ test (ASEPotTest.cpp) with reference values
• Add ci_ase.yml workflow
• Rewrite ase_pot.md with build instructions and examples

[github-actions]

eOn Documentation Preview

Download: documentation.zip

Unzip and open index.html to view.

[github-actions]

Benchmark Results

Note

All benchmarks unchanged

	Count
⚪ Unchanged	8

8 unchanged benchmark(s)

Benchmark	Before	After	Ratio
bench_eonclient.TimeMinimizationLJCluster.peakmem_minimization_lbfgs	27.2M	27.2M	~1x
bench_eonclient.TimeMinimizationLJCluster.time_minimization_lbfgs	36.7±0ms	37.5±0ms	~1.02x
bench_eonclient.TimeNEBMorsePt.peakmem_neb	27.2M	27.2M	~1x
bench_eonclient.TimeNEBMorsePt.time_neb	511±0ms	518±0ms	~1.01x
bench_eonclient.TimePointMorsePt.peakmem_point_evaluation	27.2M	27.2M	~1x
bench_eonclient.TimePointMorsePt.time_point_evaluation	6.76±0ms	7.24±0ms	~1.07x
bench_eonclient.TimeSaddleSearchMorseDimer.peakmem_saddle_search_dimer	27.2M	27.1M	~1x
bench_eonclient.TimeSaddleSearchMorseDimer.time_saddle_search_dimer	60.6±0ms	62.7±0ms	~1.03x

Details

• Base: 9160f9da
• Head: 4581068b
• Runner: ubuntu-22.04

Raw asv-spyglass output

All benchmarks:

| Change   | Before   | After    |   Ratio | Benchmark (Parameter)                                                  |
|----------|----------|----------|---------|------------------------------------------------------------------------|
|          | 27.2M    | 27.2M    |    1    | bench_eonclient.TimeMinimizationLJCluster.peakmem_minimization_lbfgs   |
|          | 36.7±0ms | 37.5±0ms |    1.02 | bench_eonclient.TimeMinimizationLJCluster.time_minimization_lbfgs      |
|          | 27.2M    | 27.2M    |    1    | bench_eonclient.TimeNEBMorsePt.peakmem_neb                             |
|          | 511±0ms  | 518±0ms  |    1.01 | bench_eonclient.TimeNEBMorsePt.time_neb                                |
|          | 27.2M    | 27.2M    |    1    | bench_eonclient.TimePointMorsePt.peakmem_point_evaluation              |
|          | 6.76±0ms | 7.24±0ms |    1.07 | bench_eonclient.TimePointMorsePt.time_point_evaluation                 |
|          | 27.2M    | 27.1M    |    1    | bench_eonclient.TimeSaddleSearchMorseDimer.peakmem_saddle_search_dimer |
|          | 60.6±0ms | 62.7±0ms |    1.03 | bench_eonclient.TimeSaddleSearchMorseDimer.time_saddle_search_dimer    |