Apply new version of JuliaFormatter

[gdalle]

Format all code with JuliaFormatter v2.0, now based on JuliaSyntax

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 55.10204% with 22 lines in your changes missing coverage. Please review.

Project coverage is 50.72%. Comparing base (9174ccb) to head (ccdd07a).
Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
...faceZygoteExt/DifferentiationInterfaceZygoteExt.jl	0.00%	3 Missing ⚠️
...ace/ext/DifferentiationInterfaceEnzymeExt/utils.jl	0.00%	2 Missing ⚠️
...xt/DifferentiationInterfaceForwardDiffExt/utils.jl	0.00%	2 Missing ⚠️
.../ext/DifferentiationInterfaceMooncakeExt/onearg.jl	0.00%	2 Missing ⚠️
...t/DifferentiationInterfaceReverseDiffExt/twoarg.jl	0.00%	2 Missing ⚠️
...ext/DifferentiationInterfaceSymbolicsExt/onearg.jl	0.00%	2 Missing ⚠️
...ationInterface/test/Misc/DifferentiateWith/test.jl	0.00%	2 Missing ⚠️
...ifferentiationInterfaceEnzymeExt/reverse_onearg.jl	0.00%	1 Missing ⚠️
...entiationInterfaceFastDifferentiationExt/twoarg.jl	0.00%	1 Missing ⚠️
.../ext/DifferentiationInterfaceMooncakeExt/twoarg.jl	0.00%	1 Missing ⚠️
... and 4 more

❗ There is a different number of reports uploaded between BASE (9174ccb) and HEAD (ccdd07a). Click for more details.

HEAD has 32 uploads less than BASE

Flag	BASE (9174ccb)	HEAD (ccdd07a)
DIT	8	4
DI	33	5

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #589       +/-   ##
===========================================
- Coverage   97.57%   50.72%   -46.85%     
===========================================
  Files         110      100       -10     
  Lines        5227     5118      -109     
===========================================
- Hits         5100     2596     -2504     
- Misses        127     2522     +2395     

Flag	Coverage Δ
DI	31.22% <41.66%> (-67.38%)	⬇️
DIT	85.61% <92.30%> (-10.03%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[gdalle]

this automatic formatting changes the semantics, turning a positional argument into a keyword argument

[domluna]

this could be disabled with separate_kwargs_with_semicolon=false

I think this is the behaviour one would want for the option though, no? It seems if it didn't make this transformation previously it would be a bug.

[domluna]

or is it such that ::Val{B}=Val(1) is not considered a keyword argument but val::Val{B}=Val(1) is?

[gdalle]

No, previously it wasn't a bug. It was an unnamed positional argument, for which I'm only interested in the type (think of it as a Holy trait).
Besides, @MilesCranmer seems to say that the same behavior can be observed for named positional arguments, which is also problematic.
Named or unnamed, positional arguments can also have default values, as long as they come at the end of the list of positional arguments and before the semicolon. They don't need to be keyword arguments to have default values.

[MilesCranmer]

Yes. Here’s an example in my codebase:

function check_constraints(
     tree::AbstractExpressionNode,
     options::AbstractOptions,
-    maxsize::Int,
+    maxsize::Int;
     cursize::Union{Int,Nothing}=nothing,
 )::Bool

This means that doing a function call check_constraints(tree, options, maxsize, cursize) will now throw an error when previously it worked.