Add some CUDA-specific tests for EnzymeTestUtils

[kshyatt]

I'm really the dog-with-chemistry-set meme here but tried to actually test some of this and see if it works, it seems there is some stuff that isn't quite functioning?

[github-actions]

Your PR no longer requires formatting changes. Thank you for your contribution!

[github-actions]

Benchmark Results

	main	8e4127e...	main / 8e4127e...
basics/make_zero/namedtuple	0.0519 ± 0.0022 μs	0.0533 ± 0.0023 μs	0.974 ± 0.058
basics/make_zero/struct	0.274 ± 0.0059 μs	0.277 ± 0.0062 μs	0.989 ± 0.031
basics/overhead	4.03 ± 0.01 ns	4.33 ± 0.01 ns	0.93 ± 0.0032
basics/remake_zero!/namedtuple	0.222 ± 0.0079 μs	0.223 ± 0.0097 μs	0.996 ± 0.056
basics/remake_zero!/struct	0.225 ± 0.0094 μs	0.226 ± 0.01 μs	0.999 ± 0.062
fold_broadcast/multidim_sum_bcast/1D	10.2 ± 0.35 μs	10.3 ± 0.34 μs	0.998 ± 0.048
fold_broadcast/multidim_sum_bcast/2D	12.2 ± 0.28 μs	12.2 ± 0.27 μs	0.998 ± 0.032
time_to_load	1.05 ± 0.0066 s	1.06 ± 0.019 s	0.99 ± 0.019

Benchmark Plots

A plot of the benchmark results has been uploaded as an artifact at https://github.com/EnzymeAD/Enzyme.jl/actions/runs/26077506034/artifacts/7075421181.

[wsmoses]

shouldn't this be 0 for size, not 1?

[kshyatt]

Whoops yeah I think so