feat(vitest): support vitest v4

[colinaaa]

Fix: #67. Fix #66

Note that this has been tested in lynx-family/lynx-stack#2158

[codspeed-hq]

Merging this PR will degrade performance by 48.35%

⚡ 1 improved benchmark
❌ 2 (👁 2) regressed benchmarks
✅ 127 untouched benchmarks
🆕 8 new benchmarks

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
🆕	WallTime	recursive fibo 15	N/A	21.8 µs	N/A
🆕	WallTime	iterative fibo 15	N/A	312 ns	N/A
🆕	WallTime	iterative fibo 20	N/A	348 ns	N/A
🆕	WallTime	recursive fibo 20	N/A	238.8 µs	N/A
⚡	WallTime	switch 2	324 ns	288 ns	+12.5%
🆕	Simulation	iterative fibo 20	N/A	17.1 µs	N/A
🆕	Simulation	recursive fibo 15	N/A	290.5 µs	N/A
🆕	Simulation	recursive fibo 20	N/A	1.4 ms	N/A
👁	Simulation	one	239.8 µs	402.2 µs	-40.37%
👁	Simulation	fibo 15	146.6 µs	283.9 µs	-48.35%
🆕	Simulation	iterative fibo 15	N/A	16.5 µs	N/A

---

_{Comparing colinaaa:colin/0131/vitest-v4 (485c0ca) with main (635c77c)}

[adriencaccia]

@colinaaa Thanks a lot for this!
I added some stuff, and we will be ready to go!

[GuillaumeLagrange]

olgtm

[GuillaumeLagrange]

Although it seems to break something in CI regarding the introspected node: https://github.com/CodSpeedHQ/codspeed-node/actions/runs/21631310644/job/62344634938?pr=71#step:7:738