Reproducibility package for rsx: a high-performance streaming toolkit for RAD-seq sex determination

Reproducibility package for the rsx software paper: the Snakemake workflow and results tree, the rsx source at its tagged release, and the exact external input archives used for the Rana berlandieri, SEX-DETector++ and findZX analyses. The 35.15 GB literature dataset is retained unchanged from the earlier versions. It records rsx v0.2.9 and complete CUDA coverage for the per-marker association tests, the Beta-Binomial Bayes factor with its directional posterior, and the sample-PCA Gram matrix. The per-repetition A100 and RTX 5070 rows behind every reported device ratio are under results/cuda/, with the scripts that derive the manuscript's tables and crossover figure from them in scripts/, so every ratio is recomputable. Also included: CPU/GPU parity evidence, literature datasets, binding checks, versioned TOML profiles, hydrated configurations, and pre-execution reproducibility archives. This version supersedes 0.2.9-repro.1, whose PAPER_SOURCE.txt gave GitHub links for the paper and results trees; both repositories are private, so those links did not resolve and provenance is now given as tags and commits. It supersedes 0.2.9-resubmission.8 and every earlier version, which also carried the manuscript source, the rendered submission PDFs, the cover letter and the letters to the reviewers. Those do not belong in a reproducibility record: the paper is the journal's and arXiv's to publish, and correspondence is neither's. They are removed here, and PAPER_SOURCE.txt records the tag that ties the paper to these results instead. No reported number differs from the earlier versions.