Remove compat with AbstractTrees 0.3

[nathanaelbosch]

I ran into issues with SymbolicUtils.jl breaking other packages that I tried using. I just found out that this was because of another package holding back AbstractTrees.jl to 0.3. Since SymbolicUtils.jl seems not to work anymore with that version, the compat should also be removed.

[github-actions]

Benchmark Results

	master	4b93db1...	t[master]/t[4b93db1...]
overhead/acrule/a+2	0.723 ± 0.013 μs	0.702 ± 0.015 μs	1.03
overhead/acrule/a+2+b	0.709 ± 0.01 μs	0.692 ± 0.016 μs	1.02
overhead/acrule/a+b	0.243 ± 0.0079 μs	0.249 ± 0.0093 μs	0.973
overhead/acrule/noop:Int	25.3 ± 0.05 ns	25.3 ± 0.058 ns	1
overhead/acrule/noop:Sym	0.034 ± 0.0046 μs	0.0342 ± 0.0037 μs	0.996
overhead/rule/noop:Int	0.0381 ± 0.00041 μs	0.0389 ± 0.0006 μs	0.979
overhead/rule/noop:Sym	0.043 ± 0.0018 μs	0.0422 ± 0.0018 μs	1.02
overhead/rule/noop:Term	0.0435 ± 0.0015 μs	0.0423 ± 0.0017 μs	1.03
overhead/ruleset/noop:Int	0.124 ± 0.0029 μs	0.127 ± 0.0021 μs	0.981
overhead/ruleset/noop:Sym	0.136 ± 0.0063 μs	0.142 ± 0.0039 μs	0.957
overhead/ruleset/noop:Term	3.32 ± 0.11 μs	3.36 ± 0.12 μs	0.986
overhead/simplify/noop:Int	0.148 ± 0.0021 μs	0.145 ± 0.001 μs	1.03
overhead/simplify/noop:Sym	0.159 ± 0.0022 μs	0.154 ± 0.0021 μs	1.04
overhead/simplify/noop:Term	0.0378 ± 0.0019 ms	0.0389 ± 0.0022 ms	0.972
overhead/simplify/randterm (+, *):serial	0.115 ± 0.0016 s	0.117 ± 0.0031 s	0.979
overhead/simplify/randterm (+, *):thread	0.0729 ± 0.024 s	0.074 ± 0.023 s	0.985
overhead/simplify/randterm (/, *):serial	0.219 ± 0.0067 ms	0.227 ± 0.0068 ms	0.965
overhead/simplify/randterm (/, *):thread	0.25 ± 0.0078 ms	0.261 ± 0.0089 ms	0.958
overhead/substitute/a	0.0589 ± 0.0015 ms	0.0587 ± 0.0014 ms	1
overhead/substitute/a,b	0.0521 ± 0.0015 ms	0.0519 ± 0.0014 ms	1
overhead/substitute/a,b,c	16.5 ± 0.6 μs	16.4 ± 0.7 μs	1
polyform/easy_iszero	31.3 ± 1.7 μs	0.033 ± 0.0015 ms	0.949
polyform/isone	2.79 ± 0.01 ns	2.79 ± 0.01 ns	1
polyform/iszero	1.74 ± 0.039 ms	1.76 ± 0.04 ms	0.991
polyform/simplify_fractions	2.32 ± 0.052 ms	2.35 ± 0.046 ms	0.99
time_to_load	4.14 ± 0.081 s	4.09 ± 0.0066 s	1.01

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).