Test allocs

[dpo]

Needs #33

Not sure where the allocations are coming from.
At the command line, hs5 doesn't allocate.

[Copilot]

Pull Request Overview

This PR adds allocation testing for optimization problems, specifically focusing on reducing memory allocations during function evaluations. The PR introduces new test files to verify zero-allocation behavior and refactors gradient and Hessian product computations to avoid temporary vector allocations.

• Adds allocation testing infrastructure for optimization problems hs5 and hs6
• Refactors gradient and Hessian product functions to use element-wise assignments instead of vectorized operations
• Updates the NLPModel constructor to use keyword arguments for bounds and adds type parameters for better performance

Reviewed Changes

Copilot reviewed 6 out of 7 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
test/runtests.jl	Includes new allocation test file
test/allocs.jl	New test file for zero-allocation verification
test/problems/hs5.jl	Refactors gradient and hprod functions to avoid allocations
test/problems/hs6.jl	Refactors gradient function to use element-wise assignment
test/Project.toml	Adds compatibility constraints for dependencies
src/structure.jl	Updates NLPModel struct with type parameters and constructor changes

Comments suppressed due to low confidence (2)

test/allocs.jl:5

• The test_zero_allocations function is called but not defined or imported in this file. This will cause the test to fail unless test_zero_allocations is defined elsewhere and imported.

      test_zero_allocations(hs5())

test/allocs.jl:12

• The test_zero_allocations function is called but not defined or imported in this file. This will cause the test to fail unless test_zero_allocations is defined elsewhere and imported.

      test_zero_allocations(hs6())

[dpo]

julia> using ManualNLPModels, NLPModelsTest

julia> include("test/problems/hs6.jl")
hs6 (generic function with 3 methods)

julia> model = hs6();

julia> table = test_allocs_nlpmodels(model)
Dict{Symbol, Float64} with 17 entries:
  :hprod_lag!             => 32.0
  :jtprod!                => 0.0
  :jprod!                 => 0.0
  :obj                    => 0.0
  :jac_op_transpose_prod! => 16.0
  :cons!                  => 0.0
  :hess_lag_op!           => NaN
  :hess_coord!            => 0.0
  :jac_op_prod!           => 0.0
  :grad!                  => 0.0
  :hess_lag_coord!        => 32.0
  :hess_lag_op_prod!      => 32.0
  :jac_coord!             => 0.0
  :hess_structure!        => 0.0
  :hprod!                 => 0.0
  :jac_structure!         => 0.0
  :hess_op_prod!          => 0.0