MethodError in distance_to_set

[odow]

julia> using Convex, HiGHS

julia> x = Variable(2, 2)
Variable
size: (2, 2)
sign: real
vexity: affine
id: 909…059

julia> fix!(x, [1 0; 0 1])
Variable
size: (2, 2)
sign: real
vexity: constant
id: 909…059
value: [1 0; 0 1]

julia> problem = satisfy([sum(x; dims = 1) <= 1])
Problem statistics
  problem is DCP         : true
  number of variables    : 1 (4 scalar elements)
  number of constraints  : 1 (2 scalar elements)
  number of coefficients : 5
  number of atoms        : 4

Solution summary
  termination status : OPTIMIZE_NOT_CALLED
  primal status      : NO_SOLUTION
  dual status        : NO_SOLUTION

Expression graph
  satisfy
   └─ nothing
  subject to
   └─ ≤ constraint (constant)
      └─ + (constant; real)
         ├─ * (constant; real)
         │  ├─ …
         │  └─ …
         └─ Convex.NegateAtom (constant; negative)
            └─ …

julia> solve!(problem, HiGHS.Optimizer)
ERROR: distance_to_set using the distance metric MathOptInterface.Utilities.ProjectionUpperBoundDistance() for set type MathOptInterface.Nonpositives has not been implemented yet.
Stacktrace:
  [1] error(s::String)
    @ Base ./error.jl:35
  [2] distance_to_set(d::MathOptInterface.Utilities.ProjectionUpperBoundDistance, ::Matrix{Float64}, set::MathOptInterface.Nonpositives)
    @ MathOptInterface.Utilities ~/.julia/packages/MathOptInterface/gBojA/src/Utilities/distance_to_set.jl:75