Cannot input BlockDiagonal into MvNormal in Distributions.jl

[biona001]

I'm not sure if this is a bug with Distributions.jl or BlockDiagonals.jl...

I have a block diagonal matrix V where each block is explicitly wrapped by Symmetric, but I cannot input V into the MvNormal constructor in Distributions.jl.

x1 = rand(5, 5)
x2 = rand(5, 5)
V1 = Symmetric(x1'*x1)
V2 = Symmetric(x2'*x2)
V = BlockDiagonal([V1, V2])
μ = rand(10)
MvNormal(μ, V)


MethodError: Cannot `convert` an object of type 
  BlockDiagonal{Float64{},Symmetric{Float64, Matrix{Float64}}} to an object of type 
  BlockDiagonal{Float64{},UpperTriangular{Float64, Matrix{Float64}}}
Closest candidates are:
  convert(::Type{T}, ::Factorization) where T<:AbstractArray at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.6/LinearAlgebra/src/factorization.jl:58
  convert(::Type{T}, ::T) where T<:AbstractArray at abstractarray.jl:14
  convert(::Type{T}, ::T) where T at essentials.jl:205

Stacktrace:
 [1] PDMats.PDMat(mat::BlockDiagonal{Float64, Symmetric{Float64, Matrix{Float64}}}, chol::Cholesky{Float64, BlockDiagonal{Float64, UpperTriangular{Float64, Matrix{Float64}}}})
   @ PDMats ~/.julia/packages/PDMats/YXkIE/src/pdmat.jl:16
 [2] PDMats.PDMat(mat::BlockDiagonal{Float64, Symmetric{Float64, Matrix{Float64}}})
   @ PDMats ~/.julia/packages/PDMats/YXkIE/src/pdmat.jl:19
 [3] MvNormal(μ::Vector{Float64}, Σ::BlockDiagonal{Float64, Symmetric{Float64, Matrix{Float64}}})
   @ Distributions ~/.julia/packages/Distributions/N1Ika/src/multivariate/mvnormal.jl:201
 [4] top-level scope
   @ In[144]:6
 [5] eval
   @ ./boot.jl:360 [inlined]
 [6] include_string(mapexpr::typeof(REPL.softscope), mod::Module, code::String, filename::String)
   @ Base ./loading.jl:1116

I have no idea why it's trying to convert a symmetric matrix into an uppertriangular matrix.