fix: the sky meets the sea, and the places rail admits it scrolls

Two defects a photograph found and no gate could.

**The sky seam.** A phone held upright drew a hard horizontal step across the
top 10.5% of the frame — a flat duller slab above a warm gradient, meeting in
one pixel. Painting the sky dome red proved the slab was the dome itself and
everything below the line was sea; so this was never missing geometry, it was
two halves of one join disagreeing about a colour.

The sea is a flat plane, so however wide it is made the camera's own far plane
cuts it at an elevation *below* the horizontal — about 12 degrees in a plan
view of the state. atmosphere.ts has fogged it to exactly the horizon colour by
then, which is the entire mechanism by which water dissolves into sky instead
of ending in an edge. The dome at that same angle has to be the same number,
and it was not: the below-horizon darkening ramped to its full 0.82 over the
first 11.5 degrees. The sea arrived at the horizon colour and met a sky already
a fifth darker. Measured 155,107,99 against 127,88,81 across one pixel.

Holding the horizon's own colour for the first 30 degrees costs nothing the
darkening was for — a floor or an ocean covers that band whenever there is one,
and where there is not, matching is the whole point. Measured after: the worst
single-row step anywhere in the sky falls from 15/255 to 8, and the 8 is water
texture rather than a line.

render/skyHorizonSeam.test.ts parses the shader text, which is a weaker
instrument than a frame and the strongest one available with no GL context. It
asserts the one number the picture proved wrong — how far down the dome still
paints the horizon colour — and it caught a real off-by-a-hair while being
written: 0.42 is sin(24.8 deg), just inside the 25 deg the sea can reach. It
also guards the backtick-in-a-template-literal trap that has now broken this
same shader three times.

**The places rail.** Its cap is min(22rem, 42vh) and rows are variable height,
so it always landed mid-row, and a row sliced through its own text reads as a
button that failed to draw rather than as more list. markOverflow writes
data-overflow and the stylesheet fades whichever end has list behind it.

Every measurement in it is guarded, because ui/mount.ts is deliberately tested
with no DOM at all: no geometry must mean no attribute, not data-overflow=NaN.
A pixel of slack at each end, because fractional layout reports a fully-scrolled
list a fraction short and a fade that never switches off dims a row over
nothing. stylesheet.test.ts asserts the two files still speak the same four
words — nothing else connects them and the failure is silent.

1,664 + 295 tests, all gates, budget 779,789 tri / 150 draws / p95 16.7 ms.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

src/test/render/skyHorizonSeam.test.ts

@@ -0,0 +1,110 @@
+/**
+ * The join between the sea and the sky, asserted where it is decided.
+ *
+ * ## The defect
+ *
+ * A phone held upright, on the California board, drew a hard horizontal step
+ * across the top **10.5%** of the frame: a flat, duller slab above a warm
+ * gradient, meeting in one pixel. It survived every gate in the repo. It is a
+ * shader constant, so no typecheck sees it; nothing here compiles GLSL, so no
+ * test saw it; and it is above the terrain, so the performance budget and the
+ * pixel-regression baselines were both unmoved.
+ *
+ * ## What causes it
+ *
+ * The sea is a flat plane. However wide it is made — and `SEA_SPAN_MULTIPLE`
+ * puts its rim nine board spans out precisely so it never ends in a visible
+ * edge — a camera standing off above it has its own **far plane** cut the water
+ * at an elevation some degrees *below* the horizontal. In a plan view of the
+ * state that angle is around twelve degrees. `atmosphere.ts` has fogged the
+ * water to exactly the horizon colour by the time it gets there, which is the
+ * whole mechanism by which water dissolves into sky instead of ending.
+ *
+ * So the dome, at that same angle, is the other half of that join. It has to be
+ * painting the same number. It was not: the below-horizon darkening ramped to
+ * its full 0.82 over the first 11.5 degrees, so the sea arrived at the horizon
+ * colour and met a sky already a fifth darker.
+ *
+ * ## What is asserted
+ *
+ * The shader's own text, parsed. That is a weaker instrument than a rendered
+ * frame and it is the strongest one available without a GL context, so what it
+ * checks is the one number a picture proved wrong: how far below the horizontal
+ * the dome still paints the horizon's own colour. A future edit is free to
+ * change the curve, the depth, or the colour — it is not free to start
+ * darkening inside the band where the sea is still being drawn.
+ */
+import assert from "node:assert/strict";
+import { readFileSync } from "node:fs";
+import { describe, it } from "node:test";
+
+const source = readFileSync(new URL("../../engine/scenekit.ts", import.meta.url), "utf8");
+
+/**
+ * The lowest elevation, as a sine, at which a board can still be drawing sea.
+ *
+ * The camera may retreat two spans from the scene origin, and the sea's rim is
+ * nine spans out, so the shallowest sight-line to the far water is about
+ * `atan(2 / 9)` = 12.5 degrees below the horizontal. Rounded out to 25 degrees,
+ * because the number that matters is the one a *fog* far plane produces rather
+ * than the geometry's rim, and the far plane moves with the stand-off.
+ */
+const SEA_MAY_REACH_SINE = Math.sin((25 * Math.PI) / 180);
+
+describe("the sky below the horizon", () => {
+  /** The one line that decides it, isolated so a failure names the right thing. */
+  const line = /color = mix\( color, uHorizon \* [\d.]+, ([^;]+)\);/.exec(source)?.[1]?.trim();
+
+  it("still exists and is still a blend toward a darker horizon", () => {
+    assert.notEqual(line, undefined, "the below-horizon term is gone from the sky shader");
+    const factor = Number(/uHorizon \* ([\d.]+)/.exec(source)?.[1]);
+    assert.ok(factor > 0 && factor <= 1, `below-horizon sky must not be brighter: ${factor}`);
+  });
+
+  it("holds the horizon's own colour through every angle the sea can reach", () => {
+    const edges = /smoothstep\(\s*([\d.]+),\s*([\d.]+),\s*- ?height\s*\)/.exec(line ?? "");
+    assert.notEqual(
+      edges,
+      null,
+      `the below-horizon ramp is no longer a smoothstep in -height: ${line}`,
+    );
+    const start = Number(edges![1]);
+    assert.ok(
+      start > SEA_MAY_REACH_SINE,
+      `the sky starts darkening ${((Math.asin(start) * 180) / Math.PI).toFixed(1)}deg below ` +
+        `the horizontal, inside the ${((Math.asin(SEA_MAY_REACH_SINE) * 180) / Math.PI).toFixed(0)}deg ` +
+        "band where the sea is still drawn — that is the seam",
+    );
+  });
+
+  /**
+   * The fix it replaced, named so it cannot be reintroduced as a tidy-up. A
+   * `clamp(-height * k, 0, 1)` is a perfectly ordinary way to write this and it
+   * is the exact shape that shipped the seam: at any k large enough to look
+   * like a horizon it is fully dark before the sea has finished.
+   */
+  it("does not go back to a clamp that saturates in the first few degrees", () => {
+    assert.ok(
+      !/clamp\(\s*- ?height \* [\d.]+/.test(source),
+      "the below-horizon term is a clamp again; see this suite's header",
+    );
+  });
+});
+
+/**
+ * Not a shader assertion, and here because this file is where the trap is
+ * documented: the sky shader is a **template literal**, and a backtick anywhere
+ * inside it — including in a comment quoting a filename — terminates it and
+ * turns the rest of the GLSL into JavaScript. It has cost this repo three
+ * builds. `tsc` does catch it, so this is a faster and clearer message rather
+ * than a new guarantee.
+ */
+describe("the sky shader's source text", () => {
+  it("carries no backtick inside the shader template literals", () => {
+    const shaders = [...source.matchAll(/(vertexShader|fragmentShader): `([\s\S]*?)`,\n/g)];
+    assert.ok(shaders.length >= 2, "the sky shaders are no longer inline template literals");
+    for (const [, which, body] of shaders) {
+      assert.ok(!(body ?? "").includes("`"), `a backtick inside ${which} ends the literal early`);
+    }
+  });
+});