purge SkyrimNet/Mantella references from architecture docs

SkyrimNet integration is retired; references in architecture docs were
rotting. Removes "SkyrimNet" mentions from runtime-engine and
inference-and-memory architecture docs (replaced with generic "multi-agent
failure-modes" / "Mantella-class failure-modes"), and from
development-conventions ("old nimmersky workloads").

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

runtime-engine/architecture.md

@@ -172,7 +172,7 @@ driver_context(cursor_at_NPC_i, mode) = {
 
 **The `knowledge_stack` is layered**, not a single bucket. Universal world-canon, district-canon (regional), NPC_i's own primary memory, and — only if the character is in the in-between dimension — clasp memory. The retrieval layer enforces the dimensional cut; the LLM never has to reason about it. See §Local memory architecture in inference-and-memory/architecture.md for the layering, propagation policy, and clasp-as-Ring-A* privacy primitive. The classic multi-agent hallucination source ("why does Kalypso suddenly remember what Anaximander confided to Phoibe?") is *structurally foreclosed* by the geometry. Write-back goes only to NPC_i's slice; the lemniscate guarantees NPC_i is not cursor-active again for at least one full roundtrip — write-back has all the time it needs without race conditions.
 
-This converts multi-NPC dialog from an **emergent-chaos problem** into a **bounded-cast scene problem with a typed runtime**. The Mantella / SkyrimNet failure-modes are foreclosed at the architecture level rather than the prompt level:
+This converts multi-NPC dialog from an **emergent-chaos problem** into a **bounded-cast scene problem with a typed runtime**. The classic multi-agent failure-modes are foreclosed at the architecture level rather than the prompt level:
 
 - **Cast = active-slots, period.** Geographic proximity does not equal participation
 - **Cursor sequences turns deterministically.** No concurrent LLM-storm