Operational Falsification Pathways for the HLV G-Lattice: Weighted Spectral Forks, Matched Null Models, and Spiral-Time Dynamic Response

This dataset contains the complete reproducibility package for Runs A--K7 & D-D3 of the HLV G-Lattice finite-graph validation and audit study. The archive includes Colab notebooks, Python scripts, CSV outputs, Excel summaries, generated figures, metadata files, and ZIP outputs for the following simulation and audit layers: Run A: main N=1000 weighted spectral-fork baseline. Run B: three-seed robustness check at N=1000. Run C: N=1728 sparse-low scaling check. Run D: dodecahedral / golden-ratio finite point-cloud diagnostics. Run E: hard ensemble falsification test with matched-control scalar-band comparisons. Run F: kernel-MMD distribution-separation test using spectral-shape and scalar-augmented feature sets. Run G: independent audit-response layer addressing duplicate-point artefacts, genuine seed variation, corrected shell-ratio diagnostics, phi-recursion guardrails, and RPL/CSL residual-update tables. And Run H ,I ,J,K-K7,D-D3 The material is provided for independent inspection, rerun, and extension of the finite-graph and finite-point-cloud diagnostics reported in the associated HLV G-Lattice manuscript. All outputs are numerical simulation and audit data only. They should not be interpreted as evidence for physical spacetime microstructure, exact dodecahedral spacetime cells, or experimental validation of the HLV framework. The purpose of the archive is reproducibility, falsification-oriented benchmarking, residual preservation, and transparent claim-state revision.