Implement self-correcting memory architecture with critic engine

Adds:
- Critic Engine (critic.js): Validates output against correction store
- Auto-detection: Identifies uncorrected patterns (quantitative without proof, vague quantification, etc.)
- Auto-storage: New patterns immediately stored as corrections
- NocoDB Schema: Tables for corrections, preferences, episodes, decisions, validation
- Memory Service (memory-service.js): NocoDB integration layer
- Response Generator (response-generator.js): End-to-end pipeline with critic
- Correction Store: JSON-based with README documentation

Behavior:
- ALL output validated before delivery
- Quantitative claims without evidence → auto-corrected
- Corrections immediately block future occurrences
- System learns from its own mistakes

Test: node architecture/test-critic.js

memory/corrections/README.md

@@ -0,0 +1,50 @@
+# Self-Correcting Memory - Correction Store
+
+## What This Is
+
+This directory contains **corrections** - patterns that identify mistakes and enforce quality.
+
+## How It Works
+
+1. **Critic Engine** validates ALL output before delivery
+2. **If violation found** → Output blocked, user notified
+3. **If new pattern detected** → Auto-stored as correction
+4. **Future occurrences** → Immediately blocked
+
+## Correction Format
+
+```json
+{
+  "id": "correction_<timestamp>_<random>",
+  "pattern": "regex pattern to match",
+  "excludePattern": "regex that exempts match",
+  "severity": "error|warning|auto-correct",
+  "message": "What to tell user",
+  "suggestion": "How to fix",
+  "blocking": true,
+  "autoDetected": true,
+  "hitCount": 0,
+  "createdAt": "2026-07-04T..."
+}
+```
+
+## Active Patterns
+
+| ID | Pattern | Severity | Action |
+|----|---------|----------|--------|
+| quantitative_without_proof | `\d+\s*(%\|percent\|times)` | auto-correct | Block + require evidence |
+| vague_quantification | `\b(many\|few\|several\|some)\b` | auto-correct | Block + require specificity |
+| unverified_done | `\b(done\|complete\|finished)\b` | error | Block + require verification |
+
+## Files
+
+- `_index.json` - All corrections (master list)
+- `correction_*.json` - Individual correction files
+- `README.md` - This file
+
+## Learning
+
+The system learns from its own mistakes:
+- Every uncorrected pattern → Becomes a correction
+- No manual intervention required
+- Immediate enforcement