Add test for MOI.Silent

[blegat]

Also fixes empty! and is_empty with universal fallback and optimizer attributes.

[codecov-io]

Codecov Report

Merging #741 into master will decrease coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #741      +/-   ##
==========================================
- Coverage   94.25%   94.25%   -0.01%     
==========================================
  Files          55       55              
  Lines        6148     6157       +9     
==========================================
+ Hits         5795     5803       +8     
- Misses        353      354       +1

Impacted Files	Coverage Δ
src/Utilities/universalfallback.jl	97.54% <ø> (-0.02%)	⬇️
src/Test/UnitTests/attributes.jl	100% <100%> (ø)	⬆️
src/Utilities/mockoptimizer.jl	90.3% <100%> (-0.47%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6a4c3e1...529fb55. Read the comment docs.

[rschwarz]

So I implemented MOI.supports(::Optimizer, ::MOI.Silent) = true for SCIP.jl, but the new test is failing for me, when using the MOIB.full_bridge_optimizer on top of a CachingOptimizer.

But it works fine when testing on direct SCIP.Optimizer. Should the call be forwarded to the underlying solver automatically?

EDIT: I found that the following method is called for the CachingOptimizer (according to which):

function supports(::ModelLike, attr::Union{AbstractModelAttribute,
                                           AbstractOptimizerAttribute})
    if !is_copyable(attr)
        throw(ArgumentError("`supports` is not defined for $attr, it is only" *
                            " defined for attributes such that `is_copyable`" *
                            " returns `true`."))
    end
    return false
end

This would be strange as there is another method specifically for ::CachingOptimizer.

[blegat]

It should be fixed by #745

[rschwarz]

Yes, that does fix it, thanks!

Yesterday, I was still ignorant of the difference between AbstractModelAttribute and AbstractOptimizerAttribute, so it was a worthwhile learning experience ;-)