wip(standalone): GPU field step 2 -- far-field shapes/strata/relief in shared C++/HLSL, GPU field+mesh

Seven LOD2-8 tiles: RX 7900 XTX field+meshing 0.335 ms vs CPU generateTile 15.81 ms; repeat runs
bit-identical; CPU W1/W8 44/44 hashes unchanged. Not yet a real gain: rock/air signs still come from
CPU masks for every sample. Viewer does not draw GPU tiles yet; no whole-view timing.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MmSoF64BbRFmk9C3p1VsBJ

Standalone/Research/OwnerShapeWorld/gpu_first_port/VoxelForgeGpuMesher.hlsl

@@ -126,6 +126,73 @@ RWStructuredBuffer<uint> CellCases : register(u0);
 RWStructuredBuffer<uint> CellLocalOffsets : register(u1);
 RWStructuredBuffer<uint> GroupVertexCounts : register(u2);
 
+struct FarFieldSampleRange { uint offset; uint count; };
+StructuredBuffer<VF_FieldPrimitive> FarShapes : register(t10);
+StructuredBuffer<FarFieldSampleRange> FarSampleShapeRanges : register(t11);
+StructuredBuffer<uint> FarSampleShapeIds : register(t12);
+StructuredBuffer<uint> FarRockAirSigns : register(t13);
+StructuredBuffer<uint> FarProtectedAir : register(t14);
+RWStructuredBuffer<float> FarFieldDensity : register(u7);
+
+cbuffer FarFieldConstants : register(b1)
+{
+    float FarStrataLo;
+    float FarStrataHi;
+    float FarStrataBlend;
+    uint FarSampleCount;
+};
+
+[numthreads(64,1,1)]
+void EvaluateFarFieldCS(uint3 dispatchId : SV_DispatchThreadID)
+{
+    const uint index = dispatchId.x;
+    if (index >= FarSampleCount) return;
+    const uint x = index % SampleDim;
+    const uint y = (index / SampleDim) % SampleDim;
+    const uint z = index / (SampleDim * SampleDim);
+    const float3 position = float3(OriginX, OriginY, OriginZ)
+        + float3(int(x) - 1, int(y) - 1, int(z) - 1) * StepVoxels;
+    const FarFieldSampleRange range = FarSampleShapeRanges[index];
+    float cave = 100000.0f;
+    float overlay = -100000.0f;
+    [loop]
+    for (uint item = 0; item < range.count; ++item)
+    {
+        const VF_FieldPrimitive shape = FarShapes[FarSampleShapeIds[range.offset + item]];
+        if ((shape.flags & VF_FIELD_PRIMITIVE_CARVE) != 0)
+        {
+            if (position.x < shape.minVoxel[0] - 6 || position.x > shape.maxVoxel[0] + 6
+                || position.y < shape.minVoxel[1] - 6 || position.y > shape.maxVoxel[1] + 6
+                || position.z < shape.minVoxel[2] - 6 || position.z > shape.maxVoxel[2] + 6)
+                continue;
+            float distance = VF_FieldPrimitiveDistance(shape,
+                position.x, position.y, position.z);
+            if ((shape.flags & VF_FIELD_PRIMITIVE_HAS_FLOOR) != 0)
+            {
+                const float floorZ = VF_FieldPrimitiveFloorAt(shape, position.x, position.y);
+                distance = VF_FieldSmoothMax(distance, floorZ - position.z,
+                    shape.floorBlendRadius);
+            }
+            cave = VF_FieldSmoothMin(cave, distance, shape.blendRadius);
+        }
+        else if (position.x >= shape.minVoxel[0] && position.x <= shape.maxVoxel[0]
+            && position.y >= shape.minVoxel[1] && position.y <= shape.maxVoxel[1]
+            && position.z >= shape.minVoxel[2] && position.z <= shape.maxVoxel[2])
+        {
+            overlay = max(overlay, -VF_FieldPrimitiveDistance(shape,
+                position.x, position.y, position.z));
+        }
+    }
+    const float base = min(position.z - FarStrataLo, FarStrataHi - position.z);
+    float rockDistance = VF_FieldSmoothMin(base, max(cave, overlay), FarStrataBlend);
+    const uint protectedAir = (FarProtectedAir[index >> 5] >> (index & 31)) & 1;
+    if (protectedAir != 0) rockDistance = min(rockDistance, -0.0625f);
+    const float rawDensity = -rockDistance;
+    const uint air = (FarRockAirSigns[index >> 5] >> (index & 31)) & 1;
+    const float magnitude = max(abs(rawDensity), 0.05f);
+    FarFieldDensity[index] = air != 0 ? magnitude : -magnitude;
+}
+
 groupshared uint ScanValues[256];
 
 [numthreads(256,1,1)]