Turn RunAttributor into a compile-time preference

[vchuravy]

Attributor yet again seems to be the cause of some illegal code transformation,
and I would like my last few days of investigating #3041 (comment) back.

This PR turns running it into an preference flag (so that folks who want it don't need to remember passing an environment flag),
but turns it off by default.

[github-actions]

Benchmark Results

	main	dc3959b...	main / dc3959b...
basics/make_zero/namedtuple	0.0535 ± 0.002 μs	0.0527 ± 0.0037 μs	1.01 ± 0.081
basics/make_zero/struct	0.277 ± 0.0051 μs	0.277 ± 0.0075 μs	0.999 ± 0.033
basics/overhead	4.95 ± 0.01 ns	4.95 ± 0.001 ns	1 ± 0.002
basics/remake_zero!/namedtuple	0.221 ± 0.0093 μs	0.224 ± 0.0086 μs	0.988 ± 0.056
basics/remake_zero!/struct	0.229 ± 0.011 μs	0.227 ± 0.011 μs	1.01 ± 0.07
fold_broadcast/multidim_sum_bcast/1D	10.3 ± 0.3 μs	10.3 ± 0.3 μs	1.01 ± 0.042
fold_broadcast/multidim_sum_bcast/2D	12.1 ± 0.28 μs	10.2 ± 0.17 μs	1.19 ± 0.034
time_to_load	1.08 ± 0.015 s	1.17 ± 0.029 s	0.923 ± 0.026

Benchmark Plots

A plot of the benchmark results has been uploaded as an artifact at https://github.com/EnzymeAD/Enzyme.jl/actions/runs/25424767044/artifacts/6825955217.

[vchuravy]

Sigh it seems necessary for 1.10 test to pass.

[wsmoses]

is there a reason for doing this since we already have the ref option?

I agree this is nicer, but technically breaking

[vchuravy]

is there a reason for doing this since we already have the ref option?

Attributor is turned off already by default for 1.12 and as #3041 shows causes compilation issues that are really frustrating to debug in 1.11 (I spent ~4days trying to root cause that bug this week).

So yes as CI shows it sometimes does positive things, but it also continuously seems to create deeply frustrating bugs and I for one had enough of it.

[wsmoses]

oh I meant more in reference to it being a preference vs the current ref [not wrt the default or not].

we should look at the remaining failure with 1.11 when disabled. part of why its okay to turn off is ive been slowly making an "attributor lite" pass for our specific needs, though not as aggressive

[vchuravy]

oh I meant more in reference to it being a preference vs the current ref [not wrt the default or not].

I think that's fine. We are promoting it from an internal interface, to something more official and documented.