feat: complete vendor repos, add edge intelligence and WASM modules

- Add 154 missing vendor files (gitignore was filtering them)
  - vendor/midstream: 564 files (was 561)
  - vendor/sublinear-time-solver: 1190 files (was 1039)
- Add ESP32 edge processing (ADR-039): presence, vitals, fall detection
- Add WASM programmable sensing (ADR-040/041) with wasm3 runtime
- Add firmware CI workflow (.github/workflows/firmware-ci.yml)
- Add wifi-densepose-wasm-edge crate for edge WASM modules
- Update sensing server, provision.py, UI components

Co-Authored-By: claude-flow <ruv@ruv.net>

vendor/sublinear-time-solver/dist/emergence/persistent-learning-system.d.ts

@@ -0,0 +1,103 @@
+/**
+ * Persistent Learning System
+ * Enables cross-session learning and knowledge accumulation
+ */
+export interface LearningTriple {
+    subject: string;
+    predicate: string;
+    object: string;
+    confidence: number;
+    timestamp: number;
+    sessionId: string;
+    sources: string[];
+}
+export interface SessionMemory {
+    sessionId: string;
+    startTime: number;
+    endTime?: number;
+    interactions: Interaction[];
+    discoveries: Discovery[];
+    performanceMetrics: any;
+}
+export interface Interaction {
+    timestamp: number;
+    type: string;
+    input: any;
+    output: any;
+    tools: string[];
+    success: boolean;
+}
+export interface Discovery {
+    timestamp: number;
+    type: 'pattern' | 'connection' | 'optimization' | 'insight';
+    content: any;
+    novelty: number;
+    utility: number;
+}
+export declare class PersistentLearningSystem {
+    private knowledgeBase;
+    private sessionMemory;
+    private currentSessionId;
+    private learningRate;
+    private forgettingRate;
+    private storagePath;
+    constructor(storagePath?: string);
+    /**
+     * Initialize new learning session
+     */
+    private initializeSession;
+    /**
+     * Learn from interaction results
+     */
+    learnFromInteraction(interaction: Interaction): Promise<void>;
+    /**
+     * Add knowledge triple with reinforcement learning
+     */
+    addKnowledge(triple: LearningTriple): Promise<void>;
+    /**
+     * Query learned knowledge with confidence scores
+     */
+    queryKnowledge(subject?: string, predicate?: string, object?: string): LearningTriple[];
+    /**
+     * Learn from cross-session patterns
+     */
+    analyzeHistoricalPatterns(): Promise<Discovery[]>;
+    /**
+     * Get learning recommendations based on historical data
+     */
+    getLearningRecommendations(): any[];
+    /**
+     * Apply forgetting to old, unused knowledge
+     */
+    applyForgetting(): Promise<void>;
+    /**
+     * Extract learning triples from interactions
+     */
+    private extractLearningTriples;
+    private extractPattern;
+    private detectPatterns;
+    private findTemporalPatterns;
+    private findToolPatterns;
+    private findSuccessPatterns;
+    private analyzeToolEffectiveness;
+    private findUnderutilizedCombinations;
+    private getSuccessfulPatterns;
+    private identifyWeakAreas;
+    private calculateNovelty;
+    private calculateUtility;
+    private recordDiscovery;
+    /**
+     * Persist knowledge to disk
+     */
+    private persistKnowledge;
+    /**
+     * Load persisted knowledge from disk
+     */
+    private loadPersistedKnowledge;
+    /**
+     * Get learning statistics
+     */
+    getLearningStats(): any;
+    private calculateAverageConfidence;
+    private getLastUpdateTime;
+}