feat: migrate embeddings to Candle WASM + remove semantic type inference

Major architectural changes:

1. EMBEDDINGS ENGINE (ONNX → Candle WASM):
   - Replace ONNX Runtime with Rust Candle compiled to WASM
   - Embedded model in WASM binary (no external downloads)
   - Quantized Q8 precision with <50MB memory footprint
   - Zero-download, offline-first operation
   - Same embedding quality (all-MiniLM-L6-v2)

2. REMOVE SEMANTIC TYPE INFERENCE:
   - Delete embeddedKeywordEmbeddings.ts (14MB of pre-computed embeddings)
   - Remove typeAwareQueryPlanner.ts and semanticTypeInference.ts
   - Remove VerbExactMatchSignal (uses keyword embeddings)
   - Update SmartRelationshipExtractor to 3 signals (55%/30%/15% weights)

API CHANGES (requires v7.0.0):
- Removed: inferTypes(), inferNouns(), inferVerbs(), inferIntent()
- Removed: getSemanticTypeInference(), SemanticTypeInference class
- Removed: TypeInference, SemanticTypeInferenceOptions types

Users can still use natural language queries in find() - they just
need to specify type explicitly for type-optimized searches.

PACKAGE SIZE IMPACT:
- Compressed: 90.1 MB → 86.2 MB (-4.3%)
- Uncompressed: 114.4 MB → 100.3 MB (-12%)
- ~448K lines of code removed

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

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

src/triple/TripleIntelligenceSystem.ts

@@ -18,7 +18,6 @@ import { TypeAwareHNSWIndex } from '../hnsw/typeAwareHNSWIndex.js'
 import { MetadataIndexManager } from '../utils/metadataIndex.js'
 import { Vector } from '../coreTypes.js'
 import { NounType } from '../types/graphTypes.js'
-import { getQueryPlanner, TypeAwareQueryPlan } from '../query/typeAwareQueryPlanner.js'
 
 // Triple Intelligence types
 export interface TripleQuery {
@@ -277,7 +276,6 @@ export class TripleIntelligenceSystem {
   
   /**
    * Main find method - executes Triple Intelligence queries
-   * Phase 3: Now with automatic type inference for 40% latency reduction
    */
   async find(query: TripleQuery, options?: TripleOptions): Promise<TripleResult[]> {
     const startTime = performance.now()
@@ -285,27 +283,6 @@ export class TripleIntelligenceSystem {
     // Validate query
     this.validateQuery(query)
 
-    // Phase 3: Infer types from natural language if not explicitly provided
-    let typeAwarePlan: TypeAwareQueryPlan | undefined
-    if (!query.types && (query.similar || query.like) && this.hnswIndex instanceof TypeAwareHNSWIndex) {
-      const queryText = query.similar || query.like!
-      const planner = getQueryPlanner()
-      typeAwarePlan = await planner.planQuery(queryText)
-
-      // Use inferred types if confidence is sufficient
-      if (typeAwarePlan.confidence > 0.6) {
-        query.types = typeAwarePlan.targetTypes
-
-        // Log for analytics
-        console.log(
-          `[Phase 3] Type inference: ${typeAwarePlan.routing} ` +
-          `(${typeAwarePlan.targetTypes.length} types, ` +
-          `confidence: ${(typeAwarePlan.confidence * 100).toFixed(0)}%, ` +
-          `estimated ${typeAwarePlan.estimatedSpeedup.toFixed(1)}x speedup)`
-        )
-      }
-    }
-
     // Build optimized query plan
     const plan = this.planner.buildPlan(query)
 
@@ -319,14 +296,6 @@ export class TripleIntelligenceSystem {
     const elapsed = performance.now() - startTime
     this.metrics.recordOperation('find_query', elapsed, results.length)
 
-    // Log Phase 3 performance impact
-    if (typeAwarePlan && typeAwarePlan.confidence > 0.6) {
-      console.log(
-        `[Phase 3] Query completed in ${elapsed.toFixed(2)}ms ` +
-        `(${results.length} results, ${typeAwarePlan.routing})`
-      )
-    }
-
     // ASSERT performance guarantees
     this.assertPerformance(elapsed, results.length)