Add game-simulation package with multi-run balance testing, fix stalled-pipeline trap

Adds a full simulation harness (game-simulation package) with greedy/random strategies,
36-metric diagnostics, multi-run orchestration via child processes, and a statistical
interpreter. Includes 2.3x engine performance optimizations (research bonus caching,
per-DC dirty tracking, reduced allocations in tick pipeline, single-pass loops).

Fixes a critical balance bug where training pipelines stalled on insufficient VRAM would
permanently block training slots — the engine never re-checked stalled pipelines, and the
greedy strategy didn't pre-check VRAM requirements. This caused 20-25% of seeds to get
stuck in Scale-up era. All three fixes (engine un-stalling, strategy VRAM pre-check,
stalled pipeline cancellation) bring pass rate from 75% to 100% across 20 random seeds.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

packages/game-engine/src/systems/competitorSystem.ts

@@ -38,17 +38,18 @@ export function processCompetitors(state: GameState): CompetitorState {
 
     const updated = { ...rival };
 
-    // Freshness decay each tick
     updated.modelFreshness = Math.max(0, updated.modelFreshness - FRESHNESS_DECAY_RATE);
 
-    // Developer ecosystem growth based on personality
     const ecoGrowth = rival.personality.openSourceTendency * 0.1 + rival.personality.marketingFocus * 0.05;
     updated.developerEcosystemScore = Math.min(100,
       updated.developerEcosystemScore + ecoGrowth * 0.01,
     );
 
-    // Catch-up: if any market share < threshold, cut prices
-    const minShare = Math.min(...Object.values(updated.marketShares));
+    const shares = Object.values(updated.marketShares);
+    let minShare = shares[0];
+    for (let i = 1; i < shares.length; i++) {
+      if (shares[i] < minShare) minShare = shares[i];
+    }
     if (minShare < COMPETITOR_CATCHUP_SHARE_THRESHOLD) {
       updated.pricingStrategy = {
         ...updated.pricingStrategy,
@@ -61,7 +62,6 @@ export function processCompetitors(state: GameState): CompetitorState {
       return updated;
     }
 
-    // Milestone reached — capability jump + model release
     const { personality } = rival;
     const capGrowth = (2 + personality.researchFocus * 5 + personality.riskTolerance * 3) *
       (1 + tick * 0.00005);
@@ -84,7 +84,6 @@ export function processCompetitors(state: GameState): CompetitorState {
     const modelIdx = Math.floor(updated.estimatedCapability / 10);
     updated.latestModelName = `${rival.name.split(' ')[0]}-${modelNames[Math.min(modelIdx, modelNames.length - 1)]}`;
 
-    // Model release resets freshness
     updated.modelFreshness = 1.0;
     updated.lastModelReleaseTick = tick;
 
@@ -96,11 +95,12 @@ export function processCompetitors(state: GameState): CompetitorState {
     return updated;
   });
 
-  const allCaps = [
-    ...rivals.filter(r => r.status === 'active').map(r => r.estimatedCapability),
-    state.models.bestDeployedModelScore,
-  ];
-  const industryBenchmark = allCaps.length > 0 ? Math.max(...allCaps) : 0;
+  let industryBenchmark = state.models.bestDeployedModelScore;
+  for (const r of rivals) {
+    if (r.status === 'active' && r.estimatedCapability > industryBenchmark) {
+      industryBenchmark = r.estimatedCapability;
+    }
+  }
 
   return { rivals, industryBenchmark };
 }