feat(archetypes): light transient magic glows

[zoeyrose]

Summary

• give the five transient meffect_* magic glows radius-1 light using their established blue, green, pink, purple, and yellow palette colors
• preserve every animation, timing, lifecycle, saving, spell, potion, and gameplay field
• reconcile Morliana's permanent Great Blue Crystal composition to one intentional neutral light9 provider by locally disabling both permanent blue overlays and removing all eight surrounding helpers
• refresh the authored semantic lighting review for source-only, runtime-created, and map-local instances without adding generated capture artifacts

Closes #94.

Coordinates

• current base: main@6047304e5f20ee3b95c04656a08cf74dcb0d6597
• head branch: feat/94-transient-magic-glows-main
• validated head: 467fedb63277f0019af2bc25d1b59f712706d1be
• worktree: /workspaces/atrinik/workspace/worktrees/content/issue-94-transient-magic-glows-main

Validation

• python3 tools/world_content_audit.py lights --check — 130 archetypes, 10 artifacts, 11,741 instances across 641 maps; 6,885 explicit colors, 4,996 intentional neutral, zero unreviewed
• python3 tools/validate.py — 165 tests plus lossless content, schema, catalog, contracts, syntax, licensing, and isolated runtime collection passed
• jq empty maps/light-source-review.json
• git diff --check
• ./atrinik build all --profile issue-94-transient-magic-glows --test — 113/114 tests passed; all relevant lighting, server, plugin, and runtime suites passed. The unchanged large smooth movement benchmark hit its fixed 900-second timeout, already tracked by perf(client): benchmark sustained map movement and cache recovery classic#196 and perf(client): add attributable lighting movement benchmarks classic#258.

Runtime verification

Exact isolated Classic profiles at Classic a10957c86f2df0f7fb66e40830f428f0a45b02cc and content 625ad548c394b75afcf04443275cb57cb77fd151 executed all 13 spell-table consumer IDs and all three potion branches, including successful, no-condition/no-target, cursed, beneficial, rapid, and clustered paths. Radius-1 colored effects rendered as bounded local glows and expired without residue, saturation, or visible flicker.

Morliana's Great Blue Crystal was verified in both smooth and discrete renderers: each shows one centered focal composition; discrete exposes the expected single-source square falloff with no adjacent radius-9 stack, while smooth produces the corresponding blended pool. Timed capture confirmed transient effects expire cleanly. No capture artifacts are committed.

Classic's separately reproduced supervisor/server control livelock remains tracked at atrinik/classic#262. It recurred after runtime completion both with explicitly closed Python-console sessions and in a no-console command-only topology; the wrapper correctly retained those exact leases for safe diagnosis. This is independent of the content diff and does not invalidate the completed gameplay/rendering observations.

The replacement runtime remains unavailable at the integration boundaries tracked by atrinik/atrinik#266, atrinik/atrinik#269, and atrinik/atrinik#270. The canonical main source independently passes its authored and semantic contracts and publishes the Classic target from this same immutable revision.

Evidence policy

No renderer screenshots, contact sheets, capture manifests, proof maps, or capture tooling are committed. Optional diagnostics remain ignored under build/ or deployment-owned output.

[github-actions]

🎉 This PR is included in version 4.0.0 🎉

The release is available on:

• v4.0.0
• GitHub release

Your semantic-release bot 📦🚀