Speed up extrema 3-50x (#58280)

This method is no longer an optimization over what Julia can do with the
naive definition on most (if not all) architectures.

Like #58267, I asked for a smattering of crowdsourced multi-architecture
benchmarking of this simple example:

```julia
using BenchmarkTools
A = rand(10000);
b1 = @benchmark extrema($A)
b2 = @benchmark mapreduce(x->(x,x),((min1, max1), (min2, max2))->(min(min1, min2), max(max1, max2)), $A)
println("$(Sys.CPU_NAME): $(round(median(b1).time/median(b2).time, digits=1))x faster")
```

With results:

```txt
cortex-a72: 13.2x faster
cortex-a76: 15.8x faster
neoverse-n1: 16.4x faster
neoverse-v2: 23.4x faster
a64fx: 46.5x faster

apple-m1: 54.9x faster
apple-m4*: 43.7x faster

znver2: 8.6x faster
znver4: 12.8x faster
znver5: 16.7x faster

haswell (32-bit): 3.5x faster
skylake-avx512: 7.4x faster
rocketlake: 7.8x faster
alderlake: 5.2x faster
cascadelake: 8.8x faster
cascadelake: 7.1x faster
```

The results are even more dramatic for Float32s, here on my M1:

```julia
julia> A = rand(Float32, 10000);

julia> @benchmark extrema($A)
BenchmarkTools.Trial: 10000 samples with 1 evaluation per sample.
 Range (min … max):  49.083 μs … 151.750 μs  ┊ GC (min … max): 0.00% … 0.00%
 Time  (median):     49.375 μs               ┊ GC (median):    0.00%
 Time  (mean ± σ):   49.731 μs ±   2.350 μs  ┊ GC (mean ± σ):  0.00% ± 0.00%

  ▅██▅▁       ▁▂▂       ▁▂▁                                    ▂
  ██████▇▇▇▇█▇████▇▆▆▆▇▇███▇▇▆▆▆▅▆▅▃▄▃▄▅▄▄▆▆▅▃▁▄▃▅▄▅▄▄▁▄▄▅▃▄▁▄ █
  49.1 μs       Histogram: log(frequency) by time      56.8 μs <

 Memory estimate: 0 bytes, allocs estimate: 0.

julia> @benchmark mapreduce(x->(x,x),((min1, max1), (min2, max2))->(min(min1, min2), max(max1, max2)), $A)
BenchmarkTools.Trial: 10000 samples with 191 evaluations per sample.
 Range (min … max):  524.435 ns …  1.104 μs  ┊ GC (min … max): 0.00% … 0.00%
 Time  (median):     525.089 ns              ┊ GC (median):    0.00%
 Time  (mean ± σ):   529.323 ns ± 20.876 ns  ┊ GC (mean ± σ):  0.00% ± 0.00%

  █▃      ▁ ▃▃                                                 ▁
  █████▇███▇███▇▇▇▇▇▇▇▇▅▆▆▆▆▆▅▅▄▆▃▄▄▃▅▅▄▃▅▄▄▄▅▅▅▃▅▄▄▁▄▄▅▆▄▄▅▄▅ █
  524 ns        Histogram: log(frequency) by time       609 ns <

 Memory estimate: 0 bytes, allocs estimate: 0.
```

Closes #34790, closes #31442, closes #44606.

---------

Co-authored-by: Mosè Giordano <[email protected]>

Author: mbauman

base/reduce.jl

@@ -860,14 +860,6 @@ ExtremaMap(::Type{T}) where {T} = ExtremaMap{Type{T}}(T)
 @inline (f::ExtremaMap)(x) = (y = f.f(x); (y, y))
 
 @inline _extrema_rf((min1, max1), (min2, max2)) = (min(min1, min2), max(max1, max2))
-# optimization for IEEEFloat
-function _extrema_rf(x::NTuple{2,T}, y::NTuple{2,T}) where {T<:IEEEFloat}
-    (x1, x2), (y1, y2) = x, y
-    anynan = isnan(x1)|isnan(y1)
-    z1 = ifelse(anynan, x1-y1, ifelse(signbit(x1-y1), x1, y1))
-    z2 = ifelse(anynan, x1-y1, ifelse(signbit(x2-y2), y2, x2))
-    z1, z2
-end
 
 ## findmax, findmin, argmax & argmin
 

test/numbers.jl

@@ -154,7 +154,8 @@ end
             x = unorded[i], unorded[i]
             y = unorded[j], unorded[j]
             z = Base._extrema_rf(x, y)
-            @test z === x || z === y
+            @test (z[1] === x[1] || z[1] === y[1]) &&
+                  (z[2] === x[1] || z[2] === y[1])
         end
     end
 end