Correctly restore unstable encoding corpus and create tokio runtime only once per process

[robert3005]

I'm trying to get fuzzer into happier state, these two things showed up as obviously wrong

[codspeed-hq]

Merging this PR will degrade performance by 22.94%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚡ 4 improved benchmarks
❌ 14 regressed benchmarks
✅ 1524 untouched benchmarks
⏩ 50 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	slice_empty_vortex	368.3 ns	2,657.8 ns	-86.14%
❌	Simulation	chunked_bool_canonical_into[(1000, 10)]	20.3 µs	35.7 µs	-43.24%
❌	Simulation	slice_vortex_buffer[1024]	871.4 ns	1,305.8 ns	-33.27%
❌	Simulation	slice_vortex_buffer[16384]	871.4 ns	1,305.8 ns	-33.27%
❌	Simulation	slice_vortex_buffer[2048]	871.4 ns	1,305.8 ns	-33.27%
❌	Simulation	slice_vortex_buffer[128]	871.4 ns	1,305.8 ns	-33.27%
❌	Simulation	slice_vortex_buffer[65536]	871.4 ns	1,305.8 ns	-33.27%
❌	Simulation	chunked_varbinview_canonical_into[(1000, 10)]	162.2 µs	198.1 µs	-18.09%
❌	Simulation	chunked_varbinview_into_canonical[(1000, 10)]	177.7 µs	214 µs	-16.98%
❌	Simulation	search_index_below_min_chunked	1.3 ms	1.5 ms	-13.61%
❌	Simulation	search_index_mixed_out_of_range_chunked	1.3 ms	1.5 ms	-13.3%
❌	Simulation	count_i32_clustered_nulls	47 µs	53.9 µs	-12.78%
❌	Simulation	search_index_full_range_random_chunked	1.4 ms	1.6 ms	-12.06%
❌	Simulation	chunked_varbinview_canonical_into[(100, 100)]	273.2 µs	308.8 µs	-11.53%
⚡	Simulation	take_10k_random	255.8 µs	197.7 µs	+29.39%
⚡	Simulation	take_10k_contiguous	276.3 µs	218.1 µs	+26.66%
⚡	Simulation	patched_take_10k_contiguous_patches	291 µs	232.1 µs	+25.39%
⚡	Simulation	patched_take_10k_random	303 µs	244 µs	+24.18%

Tip

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

---

_{Comparing rk/fuzzercleanup (e98480c) with develop (69ce1ed)²}

Footnotes

1. 50 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on develop (cb82828) during the generation of this report, so 69ce1ed was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩