WIP: Various tweaks to the Rewriters

[shashi]

• adds a ~MATCH slot which holds the whole expression
• Adds a way for Chain to stop after finding a matching rule-useful when you want to chain rules by priority
• fix metadata propagation in Walk
• inspect & pluck (inspect & pluck #538) is included here so this PR depends on that one.

[github-actions]

Benchmark Results

	master	1457894...	master/1457894fe467c2...
overhead/acrule/a+2	0.742 ± 0.025 μs	0.723 ± 0.016 μs	1.03
overhead/acrule/a+2+b	0.717 ± 0.021 μs	0.702 ± 0.017 μs	1.02
overhead/acrule/a+b	0.259 ± 0.007 μs	0.253 ± 0.0075 μs	1.02
overhead/acrule/noop:Int	26.2 ± 0.91 ns	25.9 ± 0.05 ns	1.01
overhead/acrule/noop:Sym	0.0342 ± 0.0059 μs	0.0371 ± 0.0058 μs	0.924
overhead/rule/noop:Int	0.0373 ± 0.00034 μs	0.0438 ± 0.0011 μs	0.851
overhead/rule/noop:Sym	0.0455 ± 0.0018 μs	0.055 ± 0.0021 μs	0.828
overhead/rule/noop:Term	0.045 ± 0.0013 μs	0.0557 ± 0.0027 μs	0.808
overhead/ruleset/noop:Int	0.127 ± 0.0025 μs	0.127 ± 0.0035 μs	1
overhead/ruleset/noop:Sym	0.144 ± 0.003 μs	0.161 ± 0.0046 μs	0.894
overhead/ruleset/noop:Term	6.81 ± 0.41 μs	7.21 ± 0.43 μs	0.946
overhead/simplify/noop:Int	0.149 ± 0.0012 μs	0.15 ± 0.0013 μs	0.994
overhead/simplify/noop:Sym	0.151 ± 0.0032 μs	0.17 ± 0.0057 μs	0.887
overhead/simplify/noop:Term	0.0436 ± 0.0025 ms	0.0454 ± 0.0028 ms	0.96
overhead/simplify/randterm (+, *):serial	0.134 ± 0.0068 s	0.136 ± 0.0059 s	0.981
overhead/simplify/randterm (+, *):thread	0.0838 ± 0.028 s	0.0808 ± 0.028 s	1.04
overhead/simplify/randterm (/, *):serial	0.264 ± 0.0093 ms	0.274 ± 0.01 ms	0.964
overhead/simplify/randterm (/, *):thread	0.318 ± 0.011 ms	0.309 ± 0.011 ms	1.03
overhead/substitute/a	0.124 ± 0.0032 ms	0.123 ± 0.0033 ms	1.01
overhead/substitute/a,b	0.105 ± 0.0029 ms	0.104 ± 0.0031 ms	1.01
overhead/substitute/a,b,c	17.8 ± 0.76 μs	17.4 ± 0.78 μs	1.02
polyform/easy_iszero	0.0529 ± 0.0023 ms	0.0574 ± 0.0026 ms	0.922
polyform/isone	3.1 ± 0.01 ns	3.1 ± 0.009 ns	1
polyform/iszero	5.5 ± 0.078 ms	2.44 ± 0.049 ms	2.26
polyform/simplify_fractions	3.36 ± 0.051 ms	2.7 ± 0.049 ms	1.24
time_to_load	4.48 ± 0.015 s	4.49 ± 0.013 s	0.997

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).

[Vaibhavdixit02]

This looks incorrect right?

[shashi]

yes