Failures for linear solve with cholesky factor and sparse vector

[blegat]

The first example works correctly, while makes me thing that this is a supported operation. The second one produces an error though and the third one even produces a vector with an incorrect index

julia> A = sparse(Matrix(I, 2, 2))
2×2 SparseMatrixCSC{Bool, Int64} with 2 stored entries:
 1  ⋅
 ⋅  1

julia> b = sparse(ones(2))
2-element SparseVector{Float64, Int64} with 2 stored entries:
  [1]  =  1.0
  [2]  =  1.0

julia> cholesky(A).L \ b
2-element SparseVector{Float64, Int64} with 2 stored entries:
  [1]  =  1.0
  [2]  =  1.0

julia> A = sparse(Matrix(I, 1, 1))
1×1 SparseMatrixCSC{Bool, Int64} with 1 stored entry:
 1

julia> b = sparse(ones(1))
1-element SparseVector{Float64, Int64} with 1 stored entry:
  [1]  =  1.0

julia> cholesky(A).L \ b
ERROR: SparseArrays.CHOLMOD.CHOLMODException("B cannot be stored in symmetric mode")
Stacktrace:
 [1] error_handler(status::Int32, file::Cstring, line::Int32, message::Cstring)
   @ SparseArrays.CHOLMOD ~/.julia/juliaup/julia-1.11.5+0.x64.linux.gnu/share/julia/stdlib/v1.11/SparseArrays/src/solvers/cholmod.jl:122
 [2] cholmod_l_spsolve(arg1::Int64, arg2::SparseArrays.CHOLMOD.Factor{…}, arg3::SparseArrays.CHOLMOD.Sparse{…}, arg4::Base.RefValue{…})
   @ SparseArrays.LibSuiteSparse ~/.julia/juliaup/julia-1.11.5+0.x64.linux.gnu/share/julia/stdlib/v1.11/SparseArrays/src/solvers/wrappers.jl:1458
 [3] spsolve(sys::Int64, F::SparseArrays.CHOLMOD.Factor{Float64, Int64}, B::SparseArrays.CHOLMOD.Sparse{Float64, Int64})
   @ SparseArrays.CHOLMOD ~/.julia/juliaup/julia-1.11.5+0.x64.linux.gnu/share/julia/stdlib/v1.11/SparseArrays/src/solvers/cholmod.jl:755
 [4] \
   @ ~/.julia/juliaup/julia-1.11.5+0.x64.linux.gnu/share/julia/stdlib/v1.11/SparseArrays/src/solvers/cholmod.jl:1803 [inlined]
 [5] \(L::SparseArrays.CHOLMOD.FactorComponent{Float64, :L, Int64}, B::SparseVector{Float64, Int64})
   @ SparseArrays.CHOLMOD ~/.julia/juliaup/julia-1.11.5+0.x64.linux.gnu/share/julia/stdlib/v1.11/SparseArrays/src/solvers/cholmod.jl:1847
 [6] top-level scope
   @ REPL[32]:1
Some type information was truncated. Use `show(err)` to see complete types.

julia> A = sparse(Matrix(I, 2, 2))
2×2 SparseMatrixCSC{Bool, Int64} with 2 stored entries:
 1  ⋅
 ⋅  1

julia> b = sparse(zeros(2))
2-element SparseVector{Float64, Int64} with 0 stored entries

julia> cholesky(A).L \ b
2-element SparseVector{Float64, Int64} with 1 stored entry:
  [1194158802]  =  0.0