refactor: Use option for reachable walk

[CathalMullan]

No description provided.

[codspeed-hq]

Merging this PR will degrade performance by 2.67%

⚡ 5 improved benchmarks
❌ 15 regressed benchmarks
✅ 188 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	wildcard_anchored[1]	1.5 µs	1.6 µs	-11.97%
❌	Simulation	dynamic_inline[1]	5.1 µs	5.5 µs	-5.98%
❌	Simulation	dynamic_depth[1000]	5.8 µs	6.1 µs	-5.38%
❌	Simulation	wildcard_nested[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	wildcard_nested_inline[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	wildcard_nested_present[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	wildcard_nested_triple[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	dynamic_rescan[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	wildcard_rescan[1]	1.5 µs	1.5 µs	-4.4%
❌	Simulation	wildcard_open[1]	6.1 µs	6.4 µs	-4.12%
❌	Simulation	wildcard_open[10]	6.3 µs	6.6 µs	-3.99%
❌	Simulation	wildcard_nested[10]	6.8 µs	7 µs	-3.76%
❌	Simulation	wildcard_nested_present[10]	6.9 µs	7.2 µs	-3.67%
❌	Simulation	wildcard_nested_inline[10]	6.9 µs	7.2 µs	-3.66%
❌	Simulation	wildcard_nested_triple[10]	7 µs	7.2 µs	-3.66%
⚡	Simulation	wildcard_competing[10]	6.4 µs	6.1 µs	+4.33%
⚡	Simulation	wildcard_depth[1]	1.6 µs	1.5 µs	+4.21%
⚡	Simulation	dynamic_depth[1]	1.6 µs	1.5 µs	+4.21%
⚡	Simulation	wildcard_backtrack[1]	1.6 µs	1.5 µs	+4.21%
⚡	Simulation	dynamic_competing[10]	6.8 µs	6.5 µs	+4.05%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing reachable-enum (ff20153) with main (d1ab29d)}