More abstract containers

[DanielDoehring]

Closes #2622

[github-actions]

Review checklist

This checklist is meant to assist creators of PRs (to let them know what reviewers will typically look for) and reviewers (to guide them in a structured review process). Items do not need to be checked explicitly for a PR to be eligible for merging.

Purpose and scope

• The PR has a single goal that is clear from the PR title and/or description.
• All code changes represent a single set of modifications that logically belong together.
• No more than 500 lines of code are changed or there is no obvious way to split the PR into multiple PRs.

Code quality

• The code can be understood easily.
• Newly introduced names for variables etc. are self-descriptive and consistent with existing naming conventions.
• There are no redundancies that can be removed by simple modularization/refactoring.
• There are no leftover debug statements or commented code sections.
• The code adheres to our conventions and style guide, and to the Julia guidelines.

Documentation

• New functions and types are documented with a docstring or top-level comment.
• Relevant publications are referenced in docstrings (see example for formatting).
• Inline comments are used to document longer or unusual code sections.
• Comments describe intent ("why?") and not just functionality ("what?").
• If the PR introduces a significant change or new feature, it is documented in NEWS.md with its PR number.

Testing

• The PR passes all tests.
• New or modified lines of code are covered by tests.
• New or modified tests run in less then 10 seconds.

Performance

• There are no type instabilities or memory allocations in performance-critical parts.
• If the PR intent is to improve performance, before/after time measurements are posted in the PR.

Verification

• The correctness of the code was verified using appropriate tests.
• If new equations/methods are added, a convergence test has been run and the results
  are posted in the PR.

Created with ❤️ by the Trixi.jl community.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.83%. Comparing base (4010727) to head (fe9944a).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2632   +/-   ##
=======================================
  Coverage   96.83%   96.83%           
=======================================
  Files         540      540           
  Lines       42885    42889    +4     
=======================================
+ Hits        41527    41531    +4     
  Misses       1358     1358           

Flag	Coverage Δ
unittests	96.83% <100.00%> (+<0.01%)	⬆️

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.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[DanielDoehring]

So this

Trixi.jl/test/test_unstructured_2d.jl

Line 31 in 82efbc7

semi32 = Trixi.trixi_adapt(Array, Float32, semi)

now crashes due to the inheritance/derivation of the unstructured containers from abstract container.
IMO, the current behaviour is actually somewhat unexpected as there are no adapt_structure functions implemented for the unstructured containers:

https://github.com/search?q=repo%3Atrixi-framework%2FTrixi.jl%20adapt_structure&type=code

(in contrast to the p4est containers)

Trixi.jl/src/solvers/dgsem_p4est/containers.jl

Lines 167 to 214 in 82efbc7

	# Manual adapt_structure since we have aliasing memory
	function Adapt.adapt_structure(to,
	elements::P4estElementContainer{NDIMS}) where {NDIMS}
	# Adapt underlying storage
	_node_coordinates = adapt(to, elements._node_coordinates)
	_jacobian_matrix = adapt(to, elements._jacobian_matrix)
	_contravariant_vectors = adapt(to, elements._contravariant_vectors)
	_inverse_jacobian = adapt(to, elements._inverse_jacobian)
	_surface_flux_values = adapt(to, elements._surface_flux_values)
	RealT = eltype(_inverse_jacobian)
	uEltype = eltype(_surface_flux_values)
	# Wrap arrays again
	node_coordinates = unsafe_wrap_or_alloc(to, _node_coordinates,
	size(elements.node_coordinates))
	jacobian_matrix = unsafe_wrap_or_alloc(to, _jacobian_matrix,
	size(elements.jacobian_matrix))
	contravariant_vectors = unsafe_wrap_or_alloc(to, _contravariant_vectors,
	size(jacobian_matrix))
	inverse_jacobian = unsafe_wrap_or_alloc(to, _inverse_jacobian,
	size(elements.inverse_jacobian))
	surface_flux_values = unsafe_wrap_or_alloc(to, _surface_flux_values,
	size(elements.surface_flux_values))
	new_type_params = (NDIMS,
	RealT,
	uEltype,
	NDIMS + 1,
	NDIMS + 2,
	NDIMS + 3,
	typeof(inverse_jacobian), # ArrayRealTNDIMSP1
	typeof(node_coordinates), # ArrayRealTNDIMSP2
	typeof(jacobian_matrix), # ArrayRealTNDIMSP3
	typeof(_node_coordinates), # VectorRealT
	typeof(surface_flux_values), # ArrayuEltypeNDIMSP2
	typeof(_surface_flux_values)) # VectoruEltype
	return P4estElementContainer{new_type_params...}(node_coordinates,
	jacobian_matrix,
	contravariant_vectors,
	inverse_jacobian,
	surface_flux_values,
	_node_coordinates,
	_jacobian_matrix,
	_contravariant_vectors,
	_inverse_jacobian,
	_surface_flux_values)
	end

[ranocha]

@vchuravy Could you please have a look at this and the comment of Daniel above on the adapt tests? You introduced these tests in #2212

[benegee]

I think @DanielDoehring's analysis is correct.

This fall-through

Trixi.jl/src/auxiliary/containers.jl

Lines 318 to 325 in e6496e2

	# Trixi storage types must implement these two Adapt.jl methods
	function Adapt.adapt_structure(to, c::AbstractContainer)
	error("Interface: Must implement Adapt.adapt_structure(to, ::$(typeof(c)))")
	end
	function Adapt.parent_type(C::Type{<:AbstractContainer})
	error("Interface: Must implement Adapt.parent_type(::Type{$C}")
	end

tells users to implement Adapt methods in case they want to use Adapt for their solver/mesh.

So far, only P4estMesh has Adapt methods, so either this test is supposed to fail or the fall-through is not required because Adapt itself can already figure out what to with the struct.