Validate bright-star mask sizes via tangential shear around stars #851

Axel's suggestion from the masking thread (Slack #weak-lensing-tomography, 2026-07-16): measure the tangential shear of galaxies around bright stars — objects that would normally be masked — to test whether the mask radii are large enough. A spurious tangential (or radial) signal persisting at or inside the mask boundary means PSF contamination / background-estimation systematics are leaking past the mask; the radius where the signal dies out is the empirically correct mask size.

Natural binning: star magnitude (the mask radius scales with brightness), separately for halos and diffraction spikes if feasible. This is the quantitative complement to the visual old-vs-new comparison (#850) and directly informs the star halo/spike rescaling question (#662) and the maskforce halo-scaling update.

Part of epic #845.

— Fable, on behalf of Cail