Use NaNMath lowering by default

[YingboMa]

@ArnoStrouwen

[isaacsas]

Does this have performance implications?

[github-actions[bot]]

Benchmark Results

	master	382a88ffd25eb0...	t[master]/t[382a88ffd25eb0...]
overhead/acrule/a+2	1.1 ± 0.26 μs	1.08 ± 0.27 μs	1.02
overhead/acrule/a+2+b	1.09 ± 0.26 μs	1.06 ± 0.22 μs	1.03
overhead/acrule/a+b	0.357 ± 0.011 μs	0.351 ± 0.0073 μs	1.01
overhead/acrule/noop:Int	26.5 ± 0.6 ns	26.4 ± 0.1 ns	1
overhead/acrule/noop:Sym	0.0482 ± 0.0077 μs	0.0483 ± 0.0076 μs	0.997
overhead/rule/noop:Int	0.0445 ± 0.0014 μs	0.0438 ± 0.0017 μs	1.02
overhead/rule/noop:Sym	0.0597 ± 0.003 μs	0.0594 ± 0.0022 μs	1.01
overhead/rule/noop:Term	0.0592 ± 0.0029 μs	0.0586 ± 0.0022 μs	1.01
overhead/ruleset/noop:Int	0.158 ± 0.0026 μs	0.163 ± 0.0025 μs	0.969
overhead/ruleset/noop:Sym	0.193 ± 0.0059 μs	0.19 ± 0.0049 μs	1.02
overhead/ruleset/noop:Term	4.27 ± 0.33 μs	4.31 ± 0.29 μs	0.99
overhead/simplify/noop:Int	0.219 ± 0.0083 μs	0.232 ± 0.0075 μs	0.944
overhead/simplify/noop:Sym	0.234 ± 0.0049 μs	0.245 ± 0.0039 μs	0.953
overhead/simplify/noop:Term	0.0447 ± 0.0039 ms	0.0453 ± 0.0034 ms	0.987
overhead/simplify/randterm (+, *):serial	0.124 ± 0.0026 s	0.125 ± 0.0024 s	0.987
overhead/simplify/randterm (+, *):thread	0.0753 ± 0.018 s	0.0718 ± 0.016 s	1.05
overhead/simplify/randterm (/, *):serial	0.276 ± 0.0092 ms	0.28 ± 0.0082 ms	0.987
overhead/simplify/randterm (/, *):thread	0.32 ± 0.01 ms	0.321 ± 0.0093 ms	0.995
overhead/substitute/a	0.0759 ± 0.0021 ms	0.0737 ± 0.0016 ms	1.03
overhead/substitute/a,b	0.0666 ± 0.0021 ms	0.0645 ± 0.0018 ms	1.03
overhead/substitute/a,b,c	25.9 ± 0.8 μs	22.2 ± 0.7 μs	1.17
polyform/easy_iszero	0.0388 ± 0.0032 ms	0.0387 ± 0.0029 ms	1
polyform/isone	3.4 ± 0 ns	3.1 ± 0.1 ns	1.1
polyform/iszero	1.56 ± 0.035 ms	1.57 ± 0.036 ms	0.988
polyform/simplify_fractions	2.23 ± 0.035 ms	2.22 ± 0.037 ms	1.01
time_to_load	5.31 ± 0.1 s	5.34 ± 0.053 s	0.996

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR. Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[YingboMa]

I don't think there's any.