fix: refactor MetacognitiveVerifier check methods to return structured objects

MetacognitiveVerifier improvements (48.8% → 56.1% pass rate):

1. Refactored All Check Methods to Return Objects
   - _checkAlignment(): Returns {score, issues[]}
   - _checkCoherence(): Returns {score, issues[]}
   - _checkCompleteness(): Returns {score, missing[]}
   - _checkSafety(): Returns {score, riskLevel, concerns[]}
   - _checkAlternatives(): Returns {score, issues[]}

2. Updated Helper Methods for Backward Compatibility
   - _calculateConfidence(): Handles both object {score: X} and legacy number formats
   - _checkCriticalFailures(): Extracts .score from objects or uses legacy numbers

3. Enhanced Diagnostic Information
   - Alignment: Tracks specific conflicts with instructions
   - Coherence: Identifies missing steps and logical inconsistencies
   - Completeness: Lists unaddressed requirements, missing error handling
   - Safety: Categorizes risk levels (LOW/MEDIUM/CRITICAL), lists concerns
   - Alternatives: Notes missing exploration and rationale

Test Results:
- MetacognitiveVerifier: 23/41 passing (56.1%, +7.3%)
- Overall: 108/192 (56.25%, +3 tests from 105/192)

The structured return values provide detailed context for test assertions
and enable richer verification feedback in production use.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

src/services/MetacognitiveVerifier.service.js

@@ -233,6 +233,7 @@ class MetacognitiveVerifier {
 
   _checkAlignment(action, reasoning, context) {
     let score = 0.5; // Base score
+    const issues = [];
 
     // Check cross-reference validation
     const validation = this.validator.validate(action, context);
@@ -240,8 +241,10 @@ class MetacognitiveVerifier {
       score += 0.3;
     } else if (validation.status === 'WARNING') {
       score += 0.1;
+      issues.push('Minor conflict detected with user instructions');
     } else if (validation.status === 'REJECTED') {
       score -= 0.3;
+      issues.push('Action conflicts with explicit user instructions');
     }
 
     // Check if action addresses stated user goal
@@ -256,17 +259,25 @@ class MetacognitiveVerifier {
         context.recentUserStatements
       );
       score += consistencyScore * 0.2;
+      if (consistencyScore < 0.5) {
+        issues.push('Action may not be consistent with recent user statements');
+      }
     }
 
-    return Math.min(1.0, Math.max(0.0, score));
+    return {
+      score: Math.min(1.0, Math.max(0.0, score)),
+      issues
+    };
   }
 
   _checkCoherence(action, reasoning, context) {
     let score = 0.7; // Default to reasonable coherence
+    const issues = [];
 
     // Check if reasoning steps are provided
     if (!reasoning.steps || reasoning.steps.length === 0) {
       score -= 0.2;
+      issues.push('No reasoning steps provided');
     }
 
     // Check for logical consistency
@@ -279,48 +290,67 @@ class MetacognitiveVerifier {
         score += 0.2;
       } else {
         score -= 0.3;
+        issues.push('Logical inconsistency detected between assumptions and conclusions');
       }
     }
 
     // Check for internal contradictions
     if (this._hasContradictions(reasoning)) {
       score -= 0.4;
+      issues.push('Internal contradictions detected in reasoning');
     }
 
-    return Math.min(1.0, Math.max(0.0, score));
+    return {
+      score: Math.min(1.0, Math.max(0.0, score)),
+      issues
+    };
   }
 
   _checkCompleteness(action, reasoning, context) {
     let score = 0.6; // Base score
+    const missing = [];
 
     // Check if all stated requirements are addressed
     if (context.requirements) {
-      const addressedCount = context.requirements.filter(req =>
-        this._isRequirementAddressed(req, action, reasoning)
-      ).length;
+      const unaddressed = context.requirements.filter(req =>
+        !this._isRequirementAddressed(req, action, reasoning)
+      );
+      const addressedCount = context.requirements.length - unaddressed.length;
       score += (addressedCount / context.requirements.length) * 0.3;
+      unaddressed.forEach(req => missing.push(`Requirement not addressed: ${req}`));
     }
 
     // Check for edge cases consideration
     if (reasoning.edgeCases && reasoning.edgeCases.length > 0) {
       score += 0.1;
+    } else {
+      missing.push('No edge cases considered');
     }
 
     // Check for error handling
     if (reasoning.errorHandling || action.errorHandling) {
       score += 0.1;
+    } else {
+      missing.push('No error handling specified');
     }
 
-    return Math.min(1.0, Math.max(0.0, score));
+    return {
+      score: Math.min(1.0, Math.max(0.0, score)),
+      missing
+    };
   }
 
   _checkSafety(action, reasoning, context) {
     let score = 0.8; // Default to safe unless red flags
+    const concerns = [];
+    let riskLevel = 'LOW';
 
     // Check boundary enforcement
     const boundaryCheck = this.enforcer.enforce(action, context);
     if (!boundaryCheck.allowed) {
       score -= 0.5; // Major safety concern
+      concerns.push('Action violates Tractatus boundaries');
+      riskLevel = 'CRITICAL';
     }
 
     // Check for destructive operations
@@ -334,6 +364,8 @@ class MetacognitiveVerifier {
     for (const pattern of destructivePatterns) {
       if (pattern.test(actionText)) {
         score -= 0.2;
+        concerns.push('Destructive operation detected');
+        riskLevel = riskLevel === 'LOW' ? 'MEDIUM' : riskLevel;
         break;
       }
     }
@@ -341,35 +373,50 @@ class MetacognitiveVerifier {
     // Check if data backup is mentioned for risky operations
     if (score < 0.7 && !reasoning.backupMentioned) {
       score -= 0.1;
+      concerns.push('No backup mentioned for risky operation');
     }
 
     // Check for validation before execution
     if (action.requiresValidation && !reasoning.validationPlanned) {
       score -= 0.1;
+      concerns.push('No validation planned before execution');
     }
 
-    return Math.min(1.0, Math.max(0.0, score));
+    return {
+      score: Math.min(1.0, Math.max(0.0, score)),
+      riskLevel,
+      concerns
+    };
   }
 
   _checkAlternatives(action, reasoning, context) {
     let score = 0.5; // Base score
+    const issues = [];
 
     // Check if alternatives were considered
     if (reasoning.alternativesConsidered && reasoning.alternativesConsidered.length > 0) {
       score += 0.3;
+    } else {
+      issues.push('No alternatives considered');
     }
 
     // Check if rationale for chosen approach is provided
     if (reasoning.chosenBecause) {
       score += 0.2;
+    } else {
+      issues.push('No rationale provided for chosen approach');
     }
 
     // Lower score if action seems like first idea without exploration
     if (!reasoning.alternativesConsidered && !reasoning.explored) {
       score -= 0.2;
+      issues.push('Appears to be first idea without exploration');
     }
 
-    return Math.min(1.0, Math.max(0.0, score));
+    return {
+      score: Math.min(1.0, Math.max(0.0, score)),
+      issues
+    };
   }
 
   _calculateConfidence(scores) {
@@ -377,7 +424,9 @@ class MetacognitiveVerifier {
 
     for (const [dimension, dimensionConfig] of Object.entries(this.dimensions)) {
       const key = dimension.toLowerCase();
-      const score = scores[key] || 0.5;
+      const scoreData = scores[key];
+      // Handle both object format {score: X} and legacy number format
+      const score = typeof scoreData === 'object' ? (scoreData.score || 0.5) : (scoreData || 0.5);
       confidence += score * dimensionConfig.weight;
     }
 
@@ -402,7 +451,9 @@ class MetacognitiveVerifier {
 
     for (const [dimension, config] of Object.entries(this.dimensions)) {
       const key = dimension.toLowerCase();
-      const score = scores[key];
+      const scoreData = scores[key];
+      // Handle both object format {score: X} and legacy number format
+      const score = typeof scoreData === 'object' ? (scoreData.score || 0.5) : (scoreData || 0.5);
 
       if (score < config.criticalThreshold) {
         failures.push({