Unify LaplaceDiffusion1D, LaplaceDiffusion2D, and LaplaceDiffusion3D BCs

[jlchan]

Closes #1418

[ranocha]

@sloede What's your take on this? If I remember correctly, we wanted to keep some dimension-agnostic code like this in each dimension-specific source file to make it easier to read, e.g.,

Trixi.jl/src/solvers/dgsem_tree/dg_1d.jl

Lines 479 to 483 in f404661

	# TODO: Taal dimension agnostic
	function calc_boundary_flux!(cache, t, boundary_condition::BoundaryConditionPeriodic,
	mesh::TreeMesh{1}, equations, surface_integral, dg::DG)
	@assert isempty(eachboundary(dg, cache))
	end

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.55%. Comparing base (b4fb1db) to head (553bcca).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1421      +/-   ##
==========================================
+ Coverage   89.52%   89.55%   +0.03%     
==========================================
  Files         448      448              
  Lines       35995    35979      -16     
==========================================
- Hits        32223    32219       -4     
+ Misses       3772     3760      -12     

Flag	Coverage Δ
unittests	89.55% <ø> (+0.03%)	⬆️

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.

[DanielDoehring]

In this case we also would like to delete the 3D stuff as well (which moved in in the mean time).

Trixi.jl/src/equations/laplace_diffusion_3d.jl

Lines 41 to 72 in f235619

	# Dirichlet-type boundary condition for use with a parabolic solver in weak form
	@inline function (boundary_condition::BoundaryConditionDirichlet)(flux_inner, u_inner,
	normal::AbstractVector,
	x, t,
	operator_type::Gradient,
	equations_parabolic::LaplaceDiffusion3D)
	return boundary_condition.boundary_value_function(x, t, equations_parabolic)
	end
	@inline function (boundary_condition::BoundaryConditionDirichlet)(flux_inner, u_inner,
	normal::AbstractVector,
	x, t,
	operator_type::Divergence,
	equations_parabolic::LaplaceDiffusion3D)
	return flux_inner
	end
	@inline function (boundary_condition::BoundaryConditionNeumann)(flux_inner, u_inner,
	normal::AbstractVector,
	x, t,
	operator_type::Divergence,
	equations_parabolic::LaplaceDiffusion3D)
	return boundary_condition.boundary_normal_flux_function(x, t, equations_parabolic)
	end
	@inline function (boundary_condition::BoundaryConditionNeumann)(flux_inner, u_inner,
	normal::AbstractVector,
	x, t,
	operator_type::Gradient,
	equations_parabolic::LaplaceDiffusion3D)
	return flux_inner
	end

[ranocha]

@sloede What's your take on this? If I remember correctly, we wanted to keep some dimension-agnostic code like this in each dimension-specific source file to make it easier to read, e.g.,

Trixi.jl/src/solvers/dgsem_tree/dg_1d.jl

Lines 479 to 483 in f404661

	# TODO: Taal dimension agnostic
	function calc_boundary_flux!(cache, t, boundary_condition::BoundaryConditionPeriodic,
	mesh::TreeMesh{1}, equations, surface_integral, dg::DG)
	@assert isempty(eachboundary(dg, cache))
	end

I would like to see a review from @sloede before merging this.

[sloede]

With the remaining issue resolved, this LGTM