Direct previous-CUDA check

A separate, source-bound previous/current CUDA pair ran the same complete six-second workload and numerical policy. Previous CUDA took 154.593 s; current CUDA took 132.839 s (1.164x). Both fresh workers started with isolated empty CuPy compilation caches. All result and full per-round quality checks pass. This single pair is a preliminary incremental observation, without repeated-trial stability claims; it is not pooled with the three CPU/CUDA pairs or multiplied by their ratio. Host background load was not reserved.

The baseline source is 1edc540881be9740db552b729e0e3d27cc0e4b1e and current source is 86d9b3d9aa4416cc2e6534a1d14971964e2df1b2. Complete workers, sources, imports, kernel identities and the launcher are bound by the local publication index. The parent archive audit validates this pair as well.

python -m tools.audit_bspline_lidar_imu_benchmark docs/assets/cuda_spatial_pruning/previous_cuda