perf: 1440p re-measured with the kit - 363 fps, ~12% cost

Closes the 0.1.5 performance task. Same route and preset as 0.1.0 (packaged
Development, Epic scalability, TSR, VSync off, no frame generation, 2560x1440):

  0.1.0                2.42 ms (414 fps)  p99 3.01
  0.1.5 (kit)          2.76 ms (363 fps)  p99 3.64

The 625 kit actors and 8 generated textures cost about 12% of frame time and still
leave roughly six times the 60 fps floor. Frame-time series committed at
Docs/perf/0.1.5-kit-walk-frametimes.csv; comparison table added to
Docs/release-checks.md.

Records the measurement trap once more: runs showing ~105 fps were two overlapping
game instances, not the build - 363 fps reproduced twice on a clean single instance.

TODO.md

@@ -20,7 +20,7 @@ The `0.1.4` kit, materials and theatre trim are signed off; these are the geomet
 - [A] **Retail neighbours (6822 Budget, 6806–6820, 6800 Lee Drug) from their own references.** **Started 2026-11-09.** Fetched the one usable reference Commons holds for this run — `Art/reference/retail-neighbours/01-lee-drug-and-neighbor.jpg` (Gb321, CC BY-SA 4.0, 1920×953) — which shows 6800 **and** its neighbour at street level; the vision pre-check confirmed shopfronts and parapets are visible. The 18-frame theatre set cannot stand in for these façades, and Commons has nothing else for 6822 or 6806–6820, so per-building references for those two remain the blocker. **Kit layout fixed while here:** bay count was `round(width / 4.96)`, which gave 6922 thirteen bays across 65.8 m (a window every 4.96 m on a 37 m block) and left bare wall under the cornice; bays are now capped at 9 with the module widened, and the storey spread fills the available height (6922 9 × 11, 6806 7 × 3, 6800 4 × 3, theatre 8 × 6).
 - [x] **Hero façade: fluting and the motto band as geometry.** **Done 2026-11-09.** Fluting: 20 grooves per column (thin shadow-tone cylinders set into the shaft, 120 pieces) instead of boolean cuts — it breaks the flat highlight at a fraction of the poly cost. Motto band: a real recess now, with a moulded surround standing 6 cm proud and the panel set 12 cm behind it, replacing the 8 cm dark stripe. Hero is 271 objects / 5 492 tris (from 148 / 3 044) and still 14.50 m. **En route it exposed a placement bug:** `ue_hero_facade_import.py` carried its own stale hero position `(-12024.5, -1450)` from before the `0.1.0` frontage fix, so running the importer alone moved the hero 13 m east and onto the north side — invisible in a render, and it dropped the reference score from 0.69 to 0.56. Placement now lives in `Tools/ue_placement.py` with span validators, asserted after every import (`span ok: X -14644..-12028`), and the spawn-view score recovered to **0.72**.
 - [ ] **Blosm blockout cleaning pass.** Roofs and facades remain untextured volumes and the `setupScript` styles stay unused; with the kit carrying the elevations, this is now cleanup rather than a second façade system.
-- [ ] **Re-measure 1440p performance** with 625 kit actors and 8 generated textures in the map. `0.1.0` measured 414 fps average, so there is headroom, but it has not been re-run since.
+- [x] **Re-measure 1440p performance** with 625 kit actors and 8 generated textures in the map. **Done 2026-11-09:** 2.76 ms average (363 fps), p99 3.64 ms at 2560×1440, Epic preset, TSR, VSync off, no frame generation — about **12 %** more frame time than `0.1.0` (2.42 ms / 414 fps) and still ~6× the 60 fps floor. Recorded in `Docs/release-checks.md`; series at `Docs/perf/0.1.5-kit-walk-frametimes.csv`. Note: runs showing ~105 fps were two overlapping game instances, not the build.
 - [ ] Tag `v0.1.5` and write `CHANGELOG.md` entry; then remove this section.
 
 ## 0.2.0 — MPFB player body, third-person camera, basic animation (weeks 7–9)