archive(cnn): move CNN v1 to cnn_v1/ subdirectory

Consolidate CNN v1 (CNNEffect) into dedicated directory:
- C++ effect: src/effects → cnn_v1/src/
- Shaders: workspaces/main/shaders/cnn → cnn_v1/shaders/
- Training: training/train_cnn.py → cnn_v1/training/
- Docs: doc/CNN*.md → cnn_v1/docs/

Updated all references:
- CMake source list
- C++ includes (relative paths: ../../cnn_v1/src/)
- Asset paths (../../cnn_v1/shaders/)
- Documentation cross-references

CNN v1 remains active in timeline. For new work, use CNN v2 with
enhanced features (7D static, storage buffer, sigmoid activation).

Tests: 34/34 passing (100%)

1 files changed, 55 insertions, 0 deletions

diff --git a/cnn_v1/shaders/cnn_layer.wgsl b/cnn_v1/shaders/cnn_layer.wgsl
new file mode 100644
index 0000000..cbd1686
--- /dev/null
+++ b/cnn_v1/shaders/cnn_layer.wgsl
@@ -0,0 +1,55 @@
+// CNN layer shader - uses modular convolution snippets
+// Supports multi-pass rendering with residual connections
+// DO NOT EDIT - Generated by train_cnn.py
+
+@group(0) @binding(0) var smplr: sampler;
+@group(0) @binding(1) var txt: texture_2d<f32>;
+
+#include "common_uniforms"
+#include "cnn_activation"
+#include "cnn_conv3x3"
+#include "cnn_conv5x5"
+#include "cnn_weights_generated"
+
+struct CNNLayerParams {
+    layer_index: i32,
+    blend_amount: f32,
+    _pad: vec2<f32>,
+};
+
+@group(0) @binding(2) var<uniform> uniforms: CommonUniforms;
+@group(0) @binding(3) var<uniform> params: CNNLayerParams;
+@group(0) @binding(4) var original_input: texture_2d<f32>;
+
+@vertex fn vs_main(@builtin(vertex_index) i: u32) -> @builtin(position) vec4<f32> {
+    var pos = array<vec2<f32>, 3>(
+        vec2<f32>(-1.0, -1.0), vec2<f32>(3.0, -1.0), vec2<f32>(-1.0, 3.0)
+    );
+    return vec4<f32>(pos[i], 0.0, 1.0);
+}
+
+@fragment fn fs_main(@builtin(position) p: vec4<f32>) -> @location(0) vec4<f32> {
+    // Match PyTorch linspace
+    let uv = (p.xy - 0.5) / (uniforms.resolution - 1.0);
+    let original_raw = textureSample(original_input, smplr, uv);
+    let original = (original_raw - 0.5) * 2.0;  // Normalize to [-1,1]
+    let gray = (dot(original_raw.rgb, vec3<f32>(0.2126, 0.7152, 0.0722)) - 0.5) * 2.0;
+    var result = vec4<f32>(0.0);
+
+    // Layer 0: 7→4 (RGBD output, normalizes [0,1] input)
+    if (params.layer_index == 0) {
+        result = cnn_conv5x5_7to4_src(txt, smplr, uv, uniforms.resolution, weights_layer0);
+        result = cnn_tanh(result);
+    }
+    else if (params.layer_index == 1) {
+        result = cnn_conv3x3_7to4(txt, smplr, uv, uniforms.resolution, gray, weights_layer1);
+        result = cnn_tanh(result);  // Keep in [-1,1]
+    }
+    else if (params.layer_index == 2) {
+        let sum = cnn_conv3x3_7to1(txt, smplr, uv, uniforms.resolution, gray, weights_layer2);
+        let gray_out = 1.0 / (1.0 + exp(-sum));  // Sigmoid activation
+        result = vec4<f32>(gray_out, gray_out, gray_out, 1.0);
+        return mix(original_raw, result, params.blend_amount); // [0,1]
+    }
+    return result;  // [-1,1]
+}