fix(orchestrator): pass enriched task description to all subagents

Verification, execution optimizer, and final review agents were receiving
the raw taskDescription instead of effectiveTaskDescription (which includes
research context). This caused them to work with outdated/wrong prompts
when regenerating plans.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/plan-orchestrator.ts

@@ -1088,10 +1088,10 @@ Check \`${caseDir}/ralph-wizard/research/result.json\` for:
       onProgress?.('synthesis', 'Synthesizing subagent outputs...');
       const synthesisResult = this.synthesizeResults(subagentResults);
 
-      // Phase 3: Verification
+      // Phase 3: Verification (use enriched task description from research)
       onProgress?.('verification', 'Running verification subagent...');
       const verificationResult = await this.runVerification(
-        taskDescription,
+        effectiveTaskDescription,
         synthesisResult.items,
         onProgress,
         onSubagent
@@ -1112,7 +1112,7 @@ Check \`${caseDir}/ralph-wizard/research/result.json\` for:
       try {
         onProgress?.('execution-optimization', 'Running execution optimizer for Claude Code...');
         executionResult = await this.runExecutionOptimizer(
-          taskDescription,
+          effectiveTaskDescription,
           planWithReviews,
           onProgress,
           onSubagent
@@ -1146,7 +1146,7 @@ Check \`${caseDir}/ralph-wizard/research/result.json\` for:
       try {
         onProgress?.('final-review', 'Running final review for holistic validation...');
         finalReviewResult = await this.runFinalReview(
-          taskDescription,
+          effectiveTaskDescription,
           optimizedPlan,
           onProgress,
           onSubagent