fix: canonicalize VarBin/FSST with >2GB buffers #5961

Performance Regression: -29.75%

⚡ 2 improved benchmarks
❌ 7 regressed benchmarks
✅ 1245 untouched benchmarks
⏩ 1254 skipped benchmarks¹

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
⚡	Simulation	`canonical_into_nullable[(10000, 10, 0.0)]`	529.8 µs	445.1 µs	+19.04%
❌	Simulation	`canonical_into_non_nullable[(10000, 100, 0.0)]`	1.9 ms	2.7 ms	-29.75%
⚡	Simulation	`canonical_into_nullable[(10000, 100, 0.0)]`	4.9 ms	4.1 ms	+19.84%
❌	Simulation	`canonical_into_non_nullable[(10000, 100, 0.1)]`	3.7 ms	4.5 ms	-18.15%
❌	Simulation	`canonical_into_non_nullable[(10000, 100, 0.01)]`	2.1 ms	2.9 ms	-27.39%
❌	Simulation	`into_canonical_non_nullable[(10000, 100, 0.1)]`	3.8 ms	4.6 ms	-17.55%
❌	Simulation	`into_canonical_non_nullable[(10000, 100, 0.0)]`	1.9 ms	2.7 ms	-29.31%
❌	Simulation	`into_canonical_non_nullable[(10000, 100, 0.01)]`	2.2 ms	3 ms	-26.7%
❌	Simulation	`into_canonical_nullable[(10000, 100, 0.0)]`	4.4 ms	5.2 ms	-15.41%

_{Comparing fsst-canonical (78cea9c) with develop (5483037)}

1254 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. ↩