Output folder, better TemplateExpression, colored printouts, switch to ProgressMeter

[MilesCranmer]

• This creates a full CHANGELOG.md based on all of the release notes. It also creates the 1.0.0 release notes giving descriptions of the various new features.
• Make TemplateExpression more robust
• Nice example with structs
• New URL Later.
• Make a proper output folder, rather than dumping halls of fame to the current dir
• Permit constraints and simplification by default Will add later if needed.
• Bump to DispatchDoctor.jl 0.4.17 for @nospecialize fix
• Set up user mailing list – send around an email that people can accept to join the google group for PySR/SymbolicRegression.jl updates Later.
• Custom complexities, custom constraints, etc for TemplateExpression Will add later if needed.
• Better default printout for TemplateExpressions - should print individual components! Cancelled this for now.

[github-actions]

Benchmark Results

	master	a084404...	master/a084404d28c8ec...
search/multithreading	25.7 ± 2 s	20.2 ± 2.7 s	1.27
search/serial	37.3 ± 0.77 s	32.4 ± 1.5 s	1.15
utils/best_of_sample	1.91 ± 1.6 μs	1.83 ± 1.6 μs	1.04
utils/check_constraints_x10	11.9 ± 3.2 μs	12 ± 3.2 μs	0.997
utils/compute_complexity_x10/Float64	2.11 ± 0.14 μs	2.06 ± 0.11 μs	1.02
utils/compute_complexity_x10/Int64	2.07 ± 0.12 μs	2.02 ± 0.11 μs	1.02
utils/compute_complexity_x10/nothing	1.46 ± 0.12 μs	1.42 ± 0.13 μs	1.03
utils/insert_random_op_x10	5.88 ± 1.8 μs	5.85 ± 1.9 μs	1.01
utils/next_generation_x100	0.416 ± 0.055 ms	0.44 ± 0.12 ms	0.945
utils/optimize_constants_x10	0.0352 ± 0.0084 s	0.0349 ± 0.0079 s	1.01
utils/randomly_rotate_tree_x10	5.34 ± 0.62 μs	5.22 ± 0.56 μs	1.02
time_to_load	1.59 ± 0.0043 s	1.71 ± 0.02 s	0.927

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).

[coveralls]

Pull Request Test Coverage Report for Build 11586190409

Details

• 260 of 274 (94.89%) changed or added relevant lines in 19 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage increased (+0.1%) to 95.478%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/Configure.jl	8	9	88.89%
src/ProgressBars.jl	21	22	95.45%
src/TemplateExpression.jl	86	87	98.85%
src/Options.jl	36	38	94.74%
src/OptionsStruct.jl	4	6	66.67%
src/Operators.jl	0	7	0.0%

Files with Coverage Reduction	New Missed Lines	%
src/SingleIteration.jl	1	96.97%

Totals
Change from base Build 11412416831:	0.1%
Covered Lines:	2914
Relevant Lines:	3052

---

💛 - Coveralls

[MilesCranmer]

I think in the example it would be nice to have an example with a literal user-defined struct getting returned from the structure function. And the fields of that struct could depend nonlinearly on the functions evolved.

[MilesCranmer]

e.g.,

struct Energy{T}
    hamiltonian::T
    lagrangian::T
end

energy(nt) = map(
    (T, V) -> Energy(T + V, T - V),
    nt.T, nt.V
)
energy(nt::NamedTuple{<:Any,Tuple{Vararg{String}}}) = (
    "(T = $(nt.T), V = $(nt.V))"
)

variable_mapping = (; T = [1], V = [2])

and then do a search based on some values of H and L!

(Maybe there should be a hasmethod check that falls back to this printing style as default?)