fix(newos4): measure predictor accuracy over paired predictions (v4.6.56)

The accuracy readout was structurally 0%: module-predictor.js:187 incremented the
denominator on every observed run while the numerator only moved when a stored
prediction existed, and predictions were stored only by the manual `predict`
command (the AI tool predict_task computed one and discarded it). Replaying the
AI loop's own call path with 6 tasks printed "Predictions made: 0, Accuracy: 0/6".

- pending predictions with a 15m TTL, consumed by threaded id -> task hash ->
  cli-only similarity >= 0.5; unpaired runs never enter the denominator
- duration scored within 30% or 3s (integer predictions vs fractional observations)
- errorPct capped at 500% so one miss cannot poison the mean error
- schema v2: v1 correct/total counters dropped on load, observed runs and the
  prediction counter kept; task text stored as an FNV-1a hash only
- tuning (confidence-threshold, min-data-points, pair-ttl) persists
- min-data-points now scales confidence (was declared, printed, settable, unread)
- AI loop registers a prediction per turn and threads its id; predict_task
  returns predictionId; durations exclude chat-lock queue wait
- status/dashboard say "n/a" instead of "N/A%", and carry paired count, mean
  error, turns error, pending and unpaired counts
- test: modules/test/test-predictor-accuracy.js (26 assertions)

Live on main.html v4.6.56: fresh session -> "Accuracy: n/a (0/0 paired)";
one paired run -> "100% (1/1 paired) - mean error 5% - turns +-0.0".

1 files changed, 35 insertions, 12 deletions

diff --git a/modules/core/module-aiassist.js b/modules/core/module-aiassist.js
index abbe9ce..83d073a 100644
--- a/modules/core/module-aiassist.js
+++ b/modules/core/module-aiassist.js
@@ -820,6 +820,7 @@ buildSystemPrompt() {
                     await new Promise(r => setTimeout(r, 50));
                 }
                 this._chatLock = true;
+                const taskStart = Date.now();   // predictor durations exclude chat-lock wait
                 this.terminal.markDirty();
                 // Auto-route context based on intent
                 try {
@@ -910,6 +911,16 @@ buildSystemPrompt() {
                     }
                 } catch (e) { /* world model not available */ }
 
+                // v4.6.56: register a prediction for THIS turn so recordResult can score it by id.
+                let turnPredictionId = null;
+                try {
+                    const predictor = this.terminal.moduleManager.modules['PredictorModule'];
+                    if (predictor && typeof predictor.notePrediction === 'function') {
+                        const tp = predictor.notePrediction(userMessage, 'loop');
+                        turnPredictionId = tp && tp.id ? tp.id : null;
+                    }
+                } catch (e) { /* predictor is best-effort */ }
+
                 const _traceTag = opts && opts.agentTag ? opts.agentTag : 'main';
                 try {
                     loopResult = await this.agenticLoop(messages, opts);
@@ -937,8 +948,9 @@ buildSystemPrompt() {
                         const predictor = this.terminal.moduleManager.modules['PredictorModule'];
                         if (predictor && typeof predictor.recordResult === 'function') {
                             const features = predictor._extractFeatures(userMessage);
-                            const duration = (Date.now() - lockStart) / 1000;
-                            predictor.recordResult(features, duration, loopResult.toolTurns, !this._cancelled);
+                            const duration = (Date.now() - taskStart) / 1000;
+                            predictor.recordResult(features, duration, loopResult.toolTurns, !this._cancelled,
+                                                   { task: userMessage, predictionId: turnPredictionId });
                         }
                     } catch (e) { /* predictor learning is best-effort */ }
                     // Record task in memory graph for direct AI conversations (not just orchestrator)
@@ -3020,16 +3032,26 @@ buildSystemPrompt() {
                             if (!taskDesc) return JSON.stringify({ error: 'task parameter required' });
                             var pFeatures = predictor._extractFeatures(taskDesc);
                             var pSimilar = predictor._findSimilar(pFeatures);
-                            var pResult = predictor._computePrediction(pSimilar, pFeatures);
+                            // v4.6.56: track the prediction so the outcome can score it.
+                            var pTracked;
+                            if (typeof predictor.notePrediction === 'function') {
+                                pTracked = predictor.notePrediction(taskDesc, 'tool');
+                            } else {
+                                var pFallback = predictor._computePrediction(pSimilar, pFeatures);
+                                pTracked = { id: null, predictedDuration: pFallback.duration,
+                                             predictedTurns: pFallback.toolTurns, confidence: pFallback.confidence };
+                            }
                             return JSON.stringify({
                                 task: taskDesc.substring(0, 80),
-                                estimatedDuration: pResult.duration + 's',
-                                estimatedToolTurns: pResult.toolTurns,
-                                confidence: Math.round(pResult.confidence * 100) + '%',
-                                similarTasks: pResult.similarCount,
-                                recommendation: pResult.confidence >= predictor._CONFIDENCE_THRESHOLD && pResult.duration > 30
-                                    ? 'Set TTL to ' + Math.ceil(pResult.duration * 1.5) + 's'
-                                    : 'No special action needed'
+                                predictionId: pTracked.id || null,
+                                estimatedDuration: pTracked.predictedDuration + 's',
+                                estimatedToolTurns: pTracked.predictedTurns,
+                                confidence: Math.round(pTracked.confidence * 100) + '%',
+                                similarTasks: pSimilar.length,
+                                recommendation: pTracked.confidence >= predictor._CONFIDENCE_THRESHOLD && pTracked.predictedDuration > 30
+                                    ? 'Set TTL to ' + Math.ceil(pTracked.predictedDuration * 1.5) + 's'
+                                    : 'No special action needed',
+                                tracked: !!pTracked.id
                             });
                         }
                         case 'world_model_query': {
@@ -4620,9 +4642,10 @@ buildSystemPrompt() {
                         if (sm && sm._latencyLog) totalErrors += sm._latencyLog.filter(function(e) { return e.error; }).length;
                     } catch (e) {}
                     var predictor = mm ? mm.modules['PredictorModule'] : null;
-                    var predictorAcc = 'N/A';
+                    var predictorAcc = 'n/a (no paired runs)';
                     if (predictor && predictor._accuracy && predictor._accuracy.total > 0) {
-                        predictorAcc = Math.round(predictor._accuracy.correct / predictor._accuracy.total * 100) + '%';
+                        predictorAcc = Math.round(predictor._accuracy.correct / predictor._accuracy.total * 100) + '% (' +
+                                       predictor._accuracy.correct + '/' + predictor._accuracy.total + ' paired)';
                     }
                     data.health = {
                         modules: data.modules.loaded + '/' + data.modules.total,