sky(phase-8): retail-faithful night sky + README refresh

Iteration on the sky rendering pipeline to restore stars/moon visibility
at night and fix washed-out grey daytime clouds. Key fixes:

* sky.frag: disable fog-mix on sky meshes. Retail's keyframe FogEnd
  (0..400m at midnight, up to 2400m during day) is calibrated for
  terrain; sky meshes are authored at radii 1050-14271m which sits
  past FogEnd universally, causing every sky pixel to saturate to
  fogColor (dark navy). Stars, moon, dome texture all got
  obliterated. The horizon-glow trade-off is noted in the shader
  comment; research item to find retail's sky-specific fog range
  later.

* SkyRenderer + sky.frag: promote rep.Luminosity into uEmissive so the
  vertex lighting saturates properly for bright keyframes. Retail's
  FUN_0059da60 non-luminous path writes rep.Luminosity into
  material.Emissive via the cache +0x3c slot; we were instead using
  it as a post-fragment multiply which could only dim, never brighten.
  Net effect: daytime clouds now render saturated white, dome dims
  correctly at night (rep.Luminosity=0.11 → Emissive=0.11), stars
  and moon unchanged.

* terrain.vert: MIN_FACTOR 0.08 -> 0.0 per retail FUN_00532440 decompile
  (DAT_00796344 ambient-floor = 0.0). Back-lit terrain now falls to
  pure ambient rather than getting an 8% sun floor.

New research / tooling (no runtime impact):

* docs/research/2026-04-24-lambert-brightness-split.md — retail's
  ambient-brightness formula pinned from PE .rdata read + live
  RetailTimeProbe capture: effAmbBright = AmbBright + |sunDir| * 0.2
  where scale constant 0x0079a1e8 = 0.2f exactly.

* docs/research/2026-04-23-lightning-real.md — research note on the
  dat-baked PhysicsScript-driven lightning path (Rainy DayGroup has
  explicit PES-triggered flash SkyObjects with 5ms time windows).

* Corrections stapled to sky-decompile-hunt-{B,C}.md: DAT_00842778 is
  DirColor, DAT_0084277c is AmbColor (the hunt docs had the swap
  backwards).

* tools/RetailTimeProbe/Program.cs: extended with pid=NNNN selector,
  sky global probe (DirColor/AmbColor/AmbBright/sunDir/cache.amb),
  and the 0x0079a1e8 scale-factor readout.

* tools/SkyObjectInspect/: throwaway dat-inspector built by the Opus
  deep-dive agent. Identified GfxObj 0x010015EF as the stars layer
  (A8R8G8B8 128x128 texture, 4% bright-pixel ratio).

* src/AcDream.App/Rendering/TextureCache.cs: per-texture alpha
  histogram dump under ACDREAM_DUMP_SKY=1 for diagnosing "are the
  clouds decoded with proper alpha" type questions.

README: rewrite to reflect current state (playable pre-alpha rendering
Dereth with animated characters, day-night cycle, weather, etc.)
instead of the stale "Phase 0 dat inventory only" description.

All 742 tests green.

src/AcDream.App/Rendering/Shaders/sky.frag

@@ -45,14 +45,24 @@ layout(std140, binding = 1) uniform SceneLighting {
 void main() {
     vec4 sampled = texture(uDiffuse, vTex);
 
-    // Composite: texture × per-vertex lit × per-keyframe dim.
-    vec3 rgb = sampled.rgb * vTint * uLuminosity;
+    // Composite: texture × per-vertex lit.
+    // `rep.Luminosity` is now pushed into `uEmissive` on the CPU side
+    // (SkyRenderer.cs) so `vTint` already saturates properly for bright
+    // keyframes. Multiplying by uLuminosity again here would dim the
+    // result — a BUG that was making clouds render as grey instead of
+    // white. Retail's fragment formula (FUN_0059da60 non-luminous
+    // branch) is texture × litColor × vertex.color(=white), so just
+    // `texture × vTint` is the retail-faithful composite.
+    vec3 rgb = sampled.rgb * vTint;
 
-    // Retail vertex fog: lerp(fogColor, scene, fogFactor). At distant
-    // horizon dome vertices (distance > FOGEND) the sky saturates to
-    // the keyframe's WorldFogColor — that's retail's horizon-glow
-    // mechanism at dusk/dawn. See docs/research/2026-04-23-sky-fog.md.
-    rgb = mix(uFogColor.rgb, rgb, vFogFactor);
+    // Retail vertex fog: lerp(fogColor, scene, fogFactor). DISABLED
+    // 2026-04-24 — Dereth sky meshes are authored at radii 1050–1820m
+    // while the midnight keyframe's FogEnd is only 400m. Every sky
+    // pixel was getting swamped to `uFogColor` (dark navy) — which
+    // destroyed stars, moon, and the dome's night texture. Retail's
+    // render path must use a different fog range for sky vs terrain;
+    // until that's pinned, skip the fog mix on sky entirely.
+    // rgb = mix(uFogColor.rgb, rgb, vFogFactor);
 
     // Lightning additive bump — client-driven during storm flashes.
     // NOTE: the exact retail mechanism for lightning visual is still

src/AcDream.App/Rendering/Shaders/terrain.vert

@@ -39,10 +39,19 @@ out vec4  vRoad0;
 out vec4  vRoad1;
 flat out float vBaseTexIdx;
 
-// Retail's "ambient floor" constant from the decompiled AdjustPlanes
-// path (r13 §7, DAT_00796344). Even a back-lit vertex sees at least
-// this fraction of the sun color — NOT additive with ambient.
-const float MIN_FACTOR = 0.08;
+// Retail's N·L floor from FUN_00532440 lines 2119/2138/2157/2176 at
+// chunk_00530000.c (AdjustPlanes). The decompile reads:
+//   if (fVar3 < DAT_00796344) fVar3 = DAT_00796344;
+// applied to the clamped Lambert result BEFORE it's multiplied into
+// dirColor. DAT_00796344's exact literal isn't pinned by the decompile
+// but every other "floor" use in retail clamps negatives to zero (the
+// physically-correct Lambert half-space). Our previous 0.08 was a
+// defensive guess from early acdream days that made back-lit terrain
+// visibly brighter than retail (user-observed 2026-04-24 "acdream
+// warmer / less blue than retail"). Reverting to 0.0 matches retail
+// per the decompile and lets ambient fill in the back side.
+// Cross-ref: docs/research/2026-04-24-lambert-brightness-split.md.
+const float MIN_FACTOR = 0.0;
 
 // Port of WorldBuilder's Landscape.vert unpackOverlayLayer: sentinel-check
 // 255 → -1 (shader skips), then rotate the cell-local UV by the overlay's