ruv 2d2b16a458 diag(proof): make hash precision configurable + CI cross-microarch sweep ... verify.py's HASH_QUANTIZATION_DECIMALS is now overridable via PROOF_HASH_DECIMALS. Finding: the determinism divergence is NOT Windows-vs-Linux — Windows and a second Linux box (ruvultra, same numpy/scipy) produce identical hashes at every precision, including ca58956c at 6 decimals. Only the GitHub Azure CI runner diverges (667eb054), i.e. a CPU-microarchitecture pocketfft/BLAS reordering (the #560 Skylake-vs-Cascade-Lake class). Temporary diagnostic sweep step prints the CI runner's hash at decimals 6..2 so we can pick the coarsest precision that collapses the microarch divergence to the common hash. Both the sweep step and the PROOF_HASH_DECIMALS plumbing are removed/finalized in the follow-up. Co-Authored-By: claude-flow <ruv@ruv.net>		2026-05-31 11:58:24 -04:00
..
cir_verify_helper.py	feat(signal): ADR-134 CSI→CIR via ISTA + NeumannSolver warm-start (#837)	2026-05-28 16:24:37 -04:00
expected_calibration_features.sha256	feat(signal): ADR-135 — empty-room baseline calibration	2026-05-28 18:57:08 -04:00
expected_cir_features.sha256	fix(proof): regenerate ADR-134 CIR witness hash after CIR fixes	2026-05-31 11:11:38 -04:00
expected_features.sha256	fix(verify): make v1 proof tolerant of unrelated .env keys + regen hash	2026-05-26 08:28:31 -04:00
generate_reference_signal.py	chore(repo): move v1/ → archive/v1/ + add archive/README.md (#430)	2026-04-25 23:07:52 -04:00
sample_csi_data.json	chore(repo): move v1/ → archive/v1/ + add archive/README.md (#430)	2026-04-25 23:07:52 -04:00
sample_csi_meta.json	chore(repo): move v1/ → archive/v1/ + add archive/README.md (#430)	2026-04-25 23:07:52 -04:00
verify.py	diag(proof): make hash precision configurable + CI cross-microarch sweep	2026-05-31 11:58:24 -04:00