Gravitational Manipulation Theory v1.5: Geometry-Resolved Screening, Slab-Native HUST Remapping, and Constraint Robustness

Gravitational Manipulation Theory (GMT) v1.5 replaces the sphere-proxy approximation of experimental geometry used through v1.4 with a geometry-native calculation. The framework solves the finite-thickness planar-slab boundary-value problem (BVP), defines the single-slab screening factor κ_slab, extends the calculation to a symmetric two-slab BVP to obtain the mutual screening response R_mutual^slab, and assembles the total geometry correction C_geom = G_HUST^slab / G_HUST^sphere. A strictly forward mapping, (α_DEF, λ) → κ_slab → R_mutual^slab → C_geom → O_HUST^pred, connects the theory parameters to the predicted HUST observable and provides the framework for remapping the published HUST-2020 bound into a slab-native constraint. GMT v1.5 further defines the geometry-induced boundary-shift metric, integrates the Cassini, Eöt–Wash, and HUST-v1.5 constraint coverage, and establishes a matched-grid protocol for re-evaluating the independent Eöt–Wash–HUST overlap. Epistemic status: this release establishes the geometry-resolved framework, solver validation, forward mapping chain, and matched-grid robustness protocol. Numerical quantities requiring the final solved BVP outputs—including κ_slab, R_mutual^slab, C_geom, the final slab-native HUST boundary, and N_overlap^v1.5—remain explicitly to be evaluated. Figures 13–15 therefore show illustrative constraint mappings pending the final solved-BVP evaluation. The v1.4 overlap count of 319 cells is retained as a reference only, not as a target. The final robustness verdict—survival, modification, or non-survival—is intentionally left unselected until the v1.5 numerical evaluation is completed. No detection is claimed.