KSampleADTest does not accept SentinelArrays.ChainedVector

[sgaure]

Even though a ChainedVector is an AbstractVector, the KSampleADTest does not accept such vectors. It can of course be circumvented with a collect on the vector first. DataFrames has a habit of using ChainedVector now and then.
This is HypothesisTests v0.11.5.

julia> typeof(a)
SentinelArrays.ChainedVector{Float64, Vector{Float64}}

julia> typeof(b)
SentinelArrays.ChainedVector{Float64, Vector{Float64}}

julia> KSampleADTest(a, b)
ERROR: MethodError: no method matching KSampleADTest(::Int64, ::Int64, ::Float64, ::Float64, ::Bool, ::Int64, ::SentinelArrays.ChainedVector{…}, ::Vector{…})
The type `KSampleADTest` exists, but no method is defined for this combination of argument types when trying to construct it.

Closest candidates are:
  KSampleADTest(::Int64, ::Int64, ::Float64, ::Float64, ::Bool, ::Int64, ::Vector{T}, ::Vector{Int64}) where T<:Real
   @ HypothesisTests ~/.julia/packages/HypothesisTests/xna7a/src/anderson_darling.jl:84
  KSampleADTest(::AbstractVector{T}...; modified, nsim) where T<:Real
   @ HypothesisTests ~/.julia/packages/HypothesisTests/xna7a/src/anderson_darling.jl:117
...

julia> ApproximateTwoSampleKSTest(a, b)
Approximate two sample Kolmogorov-Smirnov test
----------------------------------------------
Population details:
    parameter of interest:   Supremum of CDF differences
    value under h_0:         0.0
    point estimate:          0.019
...
julia> supertypes(typeof(a))
(SentinelArrays.ChainedVector{Float64, Vector{Float64}}, AbstractVector{Float64}, Any)

julia> supertypes(typeof(b))
(SentinelArrays.ChainedVector{Float64, Vector{Float64}}, AbstractVector{Float64}, Any)