CNN v2: storage buffer architecture foundation

- Add binary weight format (header + layer info + packed f16)
- New export_cnn_v2_weights.py for binary weight export
- Single cnn_v2_compute.wgsl shader with storage buffer
- Load weights in CNNv2Effect::load_weights()
- Create layer compute pipeline with 5 bindings
- Fast training config: 100 epochs, 3×3 kernels, 8→4→4 channels

Next: Complete bind group creation and multi-layer compute execution

1 files changed, 0 insertions, 0 deletions

diff --git a/checkpoints/checkpoint_epoch_25.pth b/checkpoints/checkpoint_epoch_25.pth
new file mode 100644
index 0000000..60da2f2
--- /dev/null
+++ b/checkpoints/checkpoint_epoch_25.pth