feat(simulation): relief_q — a local relief signal the deep rungs can resolve (T-1213)

The District and Quarter rungs rendered as flat colour, and the cause was not
the biome work everyone assumed. Measured on Ferrath through the production
canvas builder: at District the mean |elev_q delta| between neighbouring
gridunits is 0.02, and NOT ONE PAIR in a 1290x540 frame differs by 2.

elev_q spans 0-100 across the body's whole 8 km elevation range, so ONE STEP IS
80 METRES. A District canvas covers 2,048 m of ground, where the rolling relief
a walker navigates by is metres to tens of metres -- a fraction of a single
step. The sub-district detail IS generated (invent_primitives' scatter and
relief bands compute it) and then rounded away. Confirmed by running the
diagnostic with the octave cutoff disabled: still 0.02.

relief_q carries that same invented fine component against a scale chosen to
resolve it: 0-100 about a flat 50, RELIEF_FULL_SCALE_M = 400 m either side, so
8 m per step -- ten times finer than elev_q. elev_q keeps its body-absolute
meaning and the Atlas legend stays true.

Measured effect, elev_q vs relief_q (distinct values / mean 4-cell delta):

  Region     49 / 2.38   ->   101 / 21.86
  District   10 / 0.08   ->    35 / 0.35
  Quarter     8 / 0.02   ->    19 / 0.06

FIXED metre scale, never per-canvas normalization: the value for a piece of
ground must not depend on what else is in frame, or the same hillside changes
tone as the viewer pans. And it excludes elev_pct deliberately -- this is the
departure from the surrounding land, not height above sea level; including the
base would re-introduce the body-scale dominance that makes elev_q unusable
down here.

50 at the orbital rungs, which skip invent_primitives by design. Nothing is
lost: Global and Region still have varied elev_q (101 and 49 distinct values),
and the client takes whichever field carries signal via a max, with no
rung-name branching.

The client's ruggedness driver changes with it. It was an elev_q GRADIENT,
which cannot work across rungs -- the same 4-cell delta reads 21.86 at Region
and 0.08 at District, so any single full-scale constant either saturates one or
vanishes on the other. relief_q states relief outright, so |relief_q - 50| is
the answer directly and a fixed metre scale is immune to that by construction.

An absent plane reads FLAT, not zero -- 0 on this field means maximum relief
BELOW flat, so a payload without it would have stippled the entire map. That is
reachable: the field is #[serde(default)] so old-shape payloads decode. Two
colorize tests whose fixtures predate the plane caught it.

2004 server tests, 1838 client tests, 0 failed. clippy clean.

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

server/src/atlas/district_profile.rs

@@ -223,6 +223,30 @@ pub struct DistrictProfile {
     /// Average of `TerrainAnalysis.elev_pct` × 100 across district cells.
     pub elev_q: i32,
 
+    /// **Local** relief on a FIXED fine metre scale — 0–100 with 50 = flat,
+    /// each step [`RELIEF_FULL_SCALE_M`] / 50 metres (T-1213, 2026-08-07).
+    ///
+    /// WHY THIS IS NOT `elev_q`. `elev_q` spans the body's whole elevation
+    /// range: 0–100 across [`MAX_REGION_ELEVATION_KM`] = 8 km, so **one step is
+    /// 80 metres**. That is the right granularity for a whole-body map and
+    /// useless below it — a District canvas covers 2,048 m of ground, where the
+    /// rolling relief a walker navigates by is metres to tens of metres, i.e.
+    /// a fraction of one `elev_q` step. Measured on Ferrath through the
+    /// production canvas builder: at District the mean |elev_q delta| between
+    /// neighbouring gridunits is **0.02**, and not one pair in a 1290×540 frame
+    /// differs by 2. The sub-district detail IS generated — `scatter` and
+    /// `relief` below compute it — and then rounded away.
+    ///
+    /// So this field carries that same invented fine component against a scale
+    /// chosen to resolve it, rather than against the body. It is the signal the
+    /// deep rungs draw texture from; `elev_q` keeps its body-absolute meaning
+    /// and the Atlas legend stays true.
+    ///
+    /// Signed about 50 rather than a magnitude, because a hollow and a rise are
+    /// different ground and a consumer that wants only roughness can take
+    /// `|relief_q - 50|` — the reverse is not recoverable.
+    pub relief_q: i32,
+
     /// Ocean fraction for this district (0–100 scale, integer).
     pub ocean_fraction_q: i32,
 
@@ -722,6 +746,17 @@ pub fn derive_morphology_zone(
 /// span can replace it when body relief data is carried (T-1032).
 const MAX_REGION_ELEVATION_KM: f64 = 8.0;
 
+/// Full-scale swing of [`DistrictProfile::relief_q`], in metres either side of
+/// flat. 400 m spans everything from a gentle roll to a substantial ridge at
+/// the sub-district scale, and puts one quantization step at **8 m** — ten
+/// times finer than `elev_q`'s 80 m, which is what makes the deep rungs
+/// resolvable at all.
+///
+/// A FIXED metre scale on purpose, never per-canvas normalization: the value
+/// for a piece of ground must not depend on what else is in frame, or the same
+/// hillside would change tone as the viewer pans and the map would shimmer.
+const RELIEF_FULL_SCALE_M: f64 = 400.0;
+
 /// Climate tuning constants. `ClimateConstants::default()` holds the embedded
 /// values and is the authoritative runtime source **today**; the source-canonical
 /// `server/data/climate_constants.toml` mirrors those same values and is the file
@@ -1145,6 +1180,9 @@ const WINDOW_RELIEF_SALT: u64 = 0x5EED_C0DE;
 struct InventedPrimitives {
     slope_q: i32,
     elev_q: i32,
+    /// Local sub-district relief on the fixed [`RELIEF_FULL_SCALE_M`] scale —
+    /// see [`DistrictProfile::relief_q`].
+    relief_q: i32,
     ocean_fraction_q: i32,
 }
 
@@ -1360,8 +1398,22 @@ fn invent_primitives(
     // 0.0 at every cutoff ≥ 2,048 m (District and coarser), so this sum is
     // byte-identical to the pre-T-1162 `elev_pct + scatter` wherever the
     // cutoff discipline says it must be.
+    // The SAME invented fine component the elevation sum above folds in, kept
+    // on its own fixed metre scale instead of the body's (T-1213). `scatter`
+    // and `relief` are normalized fractions of MAX_REGION_ELEVATION_KM, so
+    // scaling by it gives metres; RELIEF_FULL_SCALE_M then maps +/-400 m onto
+    // 0..100 about a flat 50, i.e. 8 m per step against elev_q's 80 m.
+    //
+    // Deliberately excludes `elev_pct`: this is LOCAL relief, the departure
+    // from the surrounding land, not height above sea level. Including the
+    // base would re-introduce exactly the body-scale dominance that makes
+    // elev_q unusable down here.
+    let relief_m = (scatter + relief) * MAX_REGION_ELEVATION_KM * 1_000.0;
+    let relief_q = ((50.0 + (relief_m / RELIEF_FULL_SCALE_M) * 50.0).round() as i32).clamp(0, 100);
+
     InventedPrimitives {
         elev_q: (((elev_pct + scatter + relief) * 100.0).round() as i32).clamp(0, 100),
+        relief_q,
         slope_q: (((local_slope + ruggedness * (scatter.abs() + relief.abs()) + carve) * 100.0)
             .round() as i32)
             .clamp(0, 100),
@@ -1636,6 +1688,11 @@ fn build_district_profile(
     climate: &ClimateConstants,
     slope_q: i32,
     elev_q: i32,
+    // Local sub-district relief, fixed metre scale — see
+    // `DistrictProfile::relief_q`. Inert to every other field's derivation (a
+    // presentation-tier signal, not a classification input), so it is carried
+    // straight through to the output.
+    relief_q: i32,
     ocean_fraction_q: i32,
     region_baseline_c: Option<f32>,
     basin_direction: BasinDirection,
@@ -1733,6 +1790,7 @@ fn build_district_profile(
         precipitation_class,
         slope_q,
         elev_q,
+        relief_q,
         ocean_fraction_q,
         lake_margin_q,
         river_threshold,
@@ -1986,6 +2044,7 @@ fn derive_at_metres_with_riparian(
         climate,
         prims.slope_q,
         prims.elev_q,
+        prims.relief_q,
         prims.ocean_fraction_q,
         region_baseline_c,
         BasinDirection::default(),
@@ -2182,6 +2241,14 @@ pub fn derive_orbital_at_metres(
         climate,
         slope_q,
         elev_q,
+        // 50 = flat. The orbital rung skips `invent_primitives` altogether
+        // (that is the whole point of this path — envelope-only, no scatter,
+        // no warp), so no sub-district relief is computed here and reporting a
+        // neutral midpoint is the honest answer rather than a fabricated one.
+        // Nothing is lost visually: at Global and Region `elev_q` still varies
+        // properly — 101 and 49 distinct values respectively on Ferrath — and
+        // it is only below them that it flattens and `relief_q` takes over.
+        50,
         ocean_fraction_q,
         region_baseline_c,
         BasinDirection::default(),