Can't ldiv! a QR factorization with a BlockVector

This came up while trying to use an implicit solver with OrdinaryDiffEq.

A simple reproducible example:

julia> using BlockArrays, LinearAlgebra

julia> x = BlockArray(rand(3))
1-blocked 3-element BlockVector{Float64, Vector{Vector{Float64}}, Tuple{Base.OneTo{Int64}}}:
 0.46021136136833674
 0.5138272715735523
 0.581072254943535

julia> QR = qr(rand(3,3))
LinearAlgebra.QRCompactWY{Float64, Matrix{Float64}}
Q factor:
3×3 LinearAlgebra.QRCompactWYQ{Float64, Matrix{Float64}}:
 -0.682104  -0.0521373  -0.729394
 -0.477444  -0.723754    0.498224
 -0.553878   0.688086    0.468783
R factor:
3×3 Matrix{Float64}:
 -0.998175  -0.835741  -0.730764
  0.0        0.429548   0.36884
  0.0        0.0        0.0893908

julia> ldiv!(QR, x)
ERROR: Overload materialize!(::Lmul{ArrayLayouts.AdjQRCompactWYQLayout{ArrayLayouts.DenseColumnMajor, ArrayLayouts.DenseColumnMajor}})
Stacktrace:
 [1] error(s::String)
   @ Base ./error.jl:33
 [2] materialize!(M::ArrayLayouts.Lmul{ArrayLayouts.AdjQRCompactWYQLayout{ArrayLayouts.DenseColumnMajor, ArrayLayouts.DenseColumnMajor}, BlockArrays.BlockLayout{ArrayLayouts.DenseColumnMajor, ArrayLayouts.DenseColumnMajor}, Adjoint{Float64, LinearAlgebra.QRCompactWYQ{Float64, Matrix{Float64}}}, BlockVector{Float64, Vector{Vector{Float64}}, Tuple{Base.OneTo{Int64}}}})
   @ ArrayLayouts ~/.julia/packages/ArrayLayouts/3lnt1/src/factorizations.jl:127
 [3] lmul!(A::Adjoint{Float64, LinearAlgebra.QRCompactWYQ{Float64, Matrix{Float64}}}, B::BlockVector{Float64, Vector{Vector{Float64}}, Tuple{Base.OneTo{Int64}}})
   @ ArrayLayouts ~/.julia/packages/ArrayLayouts/3lnt1/src/lmul.jl:47
 [4] materialize!(L::ArrayLayouts.Ldiv{ArrayLayouts.QRCompactWYLayout{ArrayLayouts.DenseColumnMajor, ArrayLayouts.DenseColumnMajor}, BlockArrays.BlockLayout{ArrayLayouts.DenseColumnMajor, ArrayLayouts.DenseColumnMajor}, LinearAlgebra.QRCompactWY{Float64, Matrix{Float64}}, BlockVector{Float64, Vector{Vector{Float64}}, Tuple{Base.OneTo{Int64}}}})
   @ ArrayLayouts ~/.julia/packages/ArrayLayouts/3lnt1/src/factorizations.jl:25
 [5] ldiv!
   @ ~/.julia/packages/ArrayLayouts/3lnt1/src/ldiv.jl:89 [inlined]
 [6] ldiv!(A::LinearAlgebra.QRCompactWY{Float64, Matrix{Float64}}, x::BlockVector{Float64, Vector{Vector{Float64}}, Tuple{Base.OneTo{Int64}}})
   @ ArrayLayouts ~/.julia/packages/ArrayLayouts/3lnt1/src/ldiv.jl:135
 [7] top-level scope
   @ REPL[8]:1

Unfortunately I don't understand how ArrayLayouts works: can you give some pointers on what needs to be done?

Start with the Julia reproducer in the issue, then read ArrayLayouts' factorization handling at factorization.jl:127 and ldiv.jl:89 and 135. Trace the QRCompactWY and BlockVector path to identify the missing operation; done means ldiv!(QR, x) works for the shown BlockVector without the materialize! error.