bug: WASM engine extracts 26 fewer nodes than native (engine parity gap)

[carlos-alm]

Summary

The build benchmarks consistently show a divergence between native and WASM engines:

Metric	Native	WASM	Delta
Nodes (raw)	10,982	10,956	-26
Edges (raw)	20,869	20,870	+1
DB size	26.0 MB	19.1 MB	+36%

The 26-node shortfall means the WASM engine is silently missing ~0.24% of symbols. The DB size spread (57,725 vs 42,276 bytes/file) suggests the discrepancy extends beyond node count into how the schema is populated.

This divergence was already present in 3.4.0 and persists in 3.4.1.

Policy

Per CLAUDE.md: "Both engines must produce identical results. If they diverge, the less-accurate engine has a bug — fix it, don't document the gap."

Next Steps

1. Identify which 26 nodes the WASM engine misses (run both engines, diff the node lists)
2. Root-cause the extraction/resolution bug
3. Fix the WASM engine or native engine (whichever is incorrect)
4. Investigate the DB size gap (36% difference despite near-identical node/edge counts)

Found In

• PR docs: update build performance benchmarks (3.4.1) #647 (build benchmarks 3.4.1)
• Also present in 3.4.0 benchmarks