Track bug with Julia >= 1.9.2 (#104)

* FeasibilityResidual do not handle bound constrained problems

Author: tmigot

src/feasibility-residual.jl

@@ -48,7 +48,7 @@ function FeasibilityResidual(
   name = "$(nlp.meta.name)-feasres",
 ) where {T, S}
   if !equality_constrained(nlp)
-    if unconstrained(nlp)
+    if unconstrained(nlp) || bound_constrained(nlp)
       throw(ErrorException("Can't handle unconstrained problem"))
     elseif nlp isa AbstractNLSModel
       return FeasibilityResidual(SlackNLSModel(nlp), name = name)

test/allocs_test.jl

@@ -49,7 +49,7 @@ end
     ),
     map(
       x -> FeasibilityResidual(eval(Symbol(x))(), name = x * "feas"),
-      setdiff(problems, ["MGH01"]),
+      setdiff(problems, ["MGH01", "BNDROSENBROCK"]),
     ),
   )
   # jtprod! https://github.com/JuliaSmoothOptimizers/NLPModelsModifiers.jl/issues/77