Gravitational Manipulation Theory v1.7: Grid-Native Reconstruction, Symmetron Mutual Screening, and Reproducible Experimental Constraints

Abstract

Gravitational Manipulation Theory (GMT) v1.7 rebuilds the GMT experimental-constraint pipeline as a grid-native, physics-native, and reproducible framework. This release addresses two remaining provenance dependencies inherited from earlier GMT constraint analyses: an unreleased legacy grid-generation procedure and a tabulated two-body mutual-screening factor whose microscopic solver was implicit. Published Eöt–Wash 2020 and HUST-2020 bounds are frozen as versioned digitized inputs, the two-body response is recomputed using a native symmetron boundary-value problem, and the experimental reaches are reconstructed on a single declared master grid. The native symmetron two-sphere calculation gives an essentially inactive mutual-screening response, R_mutual^sym ≃ 1 over the tested range (0.997–1.024), compared with 0.64–0.96 in the legacy table. The laboratory geometry factor therefore reduces approximately to G_lab ≃ κ_s². On the canonical 400 × 400 grid, the independently covered Eöt–Wash–HUST coverage overlap is N_overlap^GN = 174 cells, with resolution-independent metrics f_overlap^GN = 0.0352 and A_overlap^log,GN = 0.627 dex². Resolution convergence and deterministic declared-input sensitivity give a converged headline f_overlap^GN ≃ 0.037 [0.021, 0.053] and A_overlap^log,GN ≃ 0.64 dex². The constraint-active exclusion overlap is evaluated separately as N_excl,overlap^GN = 39. The legacy v1.4 value of 319 cells, the v1.6 reconstruction of approximately 173 cells, and the native v1.7 value of 174 cells are category-separated reference values and are not an improvement ranking. The coverage plane is independent of the mutual-screening factor by construction; the change in R_mutual acts on the constraint-active exclusion mapping rather than on the coverage-count difference. The central result of v1.7 is structural rather than numerical: canonical GMT constraint metrics can now be regenerated from declared experimental inputs and native symmetron physics independently of the legacy pipeline. This release includes the manuscript, seven ACT I–VII figures, frozen experimental inputs, native symmetron solvers, grid-native coverage and exclusion pipelines, machine-readable canonical results, configuration files, and a one-command reproduction workflow. All results are conditional on the minimal double-well environmental-potential (W_Omega) class and on the declared digitizations, band definitions, and mapping assumptions. The results do not establish detection or confirmation of GMT, full parameter-space exclusion, experimentally demonstrated gravitational manipulation, or engineering controllability. This release is a reproducibility-focused continuation of Gravitational Manipulation Theory v1.4 and the subsequent geometry-resolved reconstruction program.