Update to PrettyTables to v3; Drop PrettyTables v2

[algebraicjuliabot]

This pull request changes the compat entry for the PrettyTables package from 2 to 2, 3.
This keeps the compat entries for earlier versions.

Note: I have not tested your package with this new compat entry.
It is your responsibility to make sure that your package tests pass before you merge this pull request.

[lukem12345]

Tests are currently passing on this PR due to the v2 fallback. https://github.com/AlgebraicJulia/Catlab.jl/actions/runs/21301574723

I've added support for PrettyTables v3 in this ACSets PR AlgebraicJulia/ACSets.jl#175 .

Similar changes from that PR should be applied around here

Catlab.jl/src/basic_sets/set_impls/TabularSet.jl

Line 50 in 6d03f35

function Base.show(io::IO, ::MIME"text/html", set::TabularSet)

It will probably be more convenient to do this after merging the ACSets PR and triggering a patch release, to unblock the compat bound for v3 on this repo. v2 support will need to be dropped on this repo.

[lukem12345]

@copilot Apply changes like AlgebraicJulia/ACSets.jl#175 to update PrettyTables support for Catlab. Drop v2 support.

[Copilot]

@lukem12345 I've opened a new pull request, #992, to work on those changes. Once the pull request is ready, I'll request review from you.

[kris-brown]

Hi Luke, looks like the AI created some idiosyncratic files - is it possible to create tests in the style of Catlab? (e.g. for starters, the in test folder in some file analogously located to the src file which is being tested)

[lukem12345]

Yeah the agent copied

Catlab.jl/test/basic_sets/FinSets.jl

Lines 25 to 31 in 6d03f35

	# Tables as sets.
	#################
	set = FinSet((x=[1,3,5], y=["a","b","c"]))
	@test getvalue(set) isa TabularSet
	@test eltype(set) == NamedTuple{(:x,:y),Tuple{Int,String}}
	@test length(set) == 3
	@test collect(set) == [(x=1, y="a"), (x=3, y="b"), (x=5, y="c")]

to some global test file to get around the ACSets update not being available via the registry.

[lukem12345]

Looks like tests pass. The benchmark script/ environment isn't resolving correctly. Is that action just supposed to run on main or after a version release? It seems fragile to a major version update of a dependency in a PR.
https://github.com/AlgebraicJulia/Catlab.jl/actions/runs/21873891737/job/63136982240#step:5:1400