test(noise): channel audit - cliffs ARE the only coarse-grid consumer (#84)

[wormeyman]

Closes item 4 of #84.

#83 left an assumption behind: that cliffs are the only consumer reading a multisample-bearing field on a grid other than 1 tile. That was never measured - which is exactly the status the cliff case itself had before #83.

The audit surface is much smaller than it looks

multisample appears in ~/GitHub/factorio-data @ 2.1.12 in exactly one place - vulcanus_basalt_lakes_multisample (planet-vulcanus-map-gen.lua:547), used only by vulcanus_elev. No other planet uses the primitive at all. So "audit every consumer of a multisample-bearing field" reduces to "audit every consumer of Vulcanus elevation":

consumer	reads elevation?	grid	how established
cliff generator	yes, via cliff_elevation_from_elevation	4	measured, #83
tile generator (19 *_range expressions)	yes, via vulcanus_elev	1	measured, here
vulcanus_temperature	yes, min(elev, elev/100)	1	same program as tiles
resources: tungsten, coal, calcite, sulfuric acid	no	n/a	read the Lua
rocks, geyser	no	n/a	read the Lua

The resource row had to be checked rather than assumed, because the generic path does couple to elevation: starting_resources_lake_mask = clamp((elevation - 1) / 10, 0, 1) feeds starting_patches' spot_favorability_expression (core/prototypes/noise-programs.lua:270), and CLAUDE.md flags that coupling as exactly what changed at 2.1.9. Vulcanus does not take that path - each of its four resources is placed by its own vulcanus_*_region expression, and those four definitions contain 0 references to elevation.

The tile generator reads grid 1, measured against the game

Substituting the cliff generator's 4-tile elevation into the tile resolver, compared against the game's own get_tile output over the 381 oracle positions:

	grid 1 (ships)	grid 4
tile-name agreement	0.9816	0.8609
lava misclassifications	0 / 381	27 / 381

46 tiles would be named wrongly, and the binary lava call - the only thing the cliff collision rejection reads - goes from exact to 27 wrong.

The sensitivity is the part worth insisting on. PR #57's field substitution failed precisely because "nothing changed" could not be distinguished from "the substitution never ran". Both arms here carry that guard explicitly, and the spec also asserts the max(-500, ...) clamp never bites at these positions, so the swap really is "the same field at grid 4" rather than something subtly different.

Side result

This clears the lava perimeter that costs 13 real cliffs their placement (#86): reading the other elevation channel makes lava dramatically worse, not better. So the perimeter error is not a channel mistake - it is somewhere else in vulcanusCatalog.

pnpm run verify green: 1310 passed / 3 skipped, 0 warnings.

🤖 Generated with Claude Code

https://claude.ai/code/session_01WRKSNgkidxc6daeHGJHqpt