Port OreSpawn 4.0.16 to NeoForge 1.20.6

[SkyBlade1978]

Summary

Ports OreSpawn 4.0.16 to Minecraft 1.20.6 on NeoForge, ending at 4.0.16.120062.

This carries the complete 4.0.8-4.0.16 campaign:

• preserves long Ore and Fluid editor values;
• safely replaces provider-declared natural terrain hosts while preserving later authored and unsafe blocks;
• evaluates Stable Layers rock ranges against actual world height;
• resolves dynamic biome weights and ore filters through stable biome keys;
• aligns public sampling with generated surface and quart-biome attribution;
• adds the existing provider biome-filter API fields and namespaced geome handling;
• converts exposed Snow and surface Ice without touching buried/authored controls;
• leaves GameTest result reporting and shutdown to its harness while retaining ordinary benchmark auto-stop.

NeoForge 1.20.6 adaptations

• Retains NeoForge 20.6.139, NeoGradle 7.1.38, Gradle 9.2.1, Java 21 and pack format 32.
• Uses NeoForge registries, event buses, biome modifiers, Common c: biome tags and META-INF/neoforge.mods.toml.
• Keeps constructor-based ResourceLocation, the existing BiomeRegistryAccess, processed-resource Eclipse launches and the target's six access-transformer rules.
• Adds a genuinely empty-cache NeoGradle bootstrap plus same-cache offline replay; no ForgeGradle, Mavenizer or Java 25 bootstrap is introduced.
• Produces deterministic version-only main, sources and Javadoc artifacts with guarded Maven preparation.

API major 1, schemas 4/6/5, registry IDs, profile filenames, migration semantics, provider field meanings, existing chunks and the [4.0.6,5.0.0) compatibility range remain unchanged.

Validation

The exact candidate 41b5e834735877309f6dd1b500a178b6aa626a9f received manual acceptance and passed:

• 135 tests across 33 suites with no failures, errors or skips;
• 39 JSON/mcmeta files, 16 locales with 359 keys and all 21 documentation exports;
• the full offline build, Javadocs, artifact audit, checksums and NeoGradle Eclipse lifecycle;
• empty-cache Java 21 bootstrap and same-cache offline replay;
• exact packaged NeoForge client/server fresh and same-save reload tests;
• natural-host, low-world height, dynamic-biome, Snow/Ice, flowing-fluid and stable-attribution gates;
• GameTest and ordinary benchmark lifecycle checks with clean logs and no surviving processes.

Matched final 81-chunk medians were Sky 45.341, Cyano 43.180 and vanilla 42.152 ms/chunk. Sky was 5.005% above Cyano and 7.565% above vanilla, below the 10% investigation threshold. The final candidate was faster than every staged 4.0.8-4.0.11 candidate.

Artifact SHA-256:

• main: C3642CE69BE10E140BCE12345FBA1AF22C06E8A1D92F34B4C1E1356027D692F0
• sources: 71157A050A8B509A2C879EE4ECE29C4F21BF5F921206897DFE14D359A4E03A17
• Javadocs: A3396FBFA70ED5FE17F941011562A3D0210A25C539C3F2E211C3463326F7A62E

This PR does not tag, publish, or directly update an MMD release branch.