inconsistent MArray allocation behavior with @inbounds, loop order

Allocations for MArray in an accumulation loop are inconsistent (depend on @inbounds and loop order)

function accum1(L, ::Val{N}) where {N}
    acc = zero(MVector{N,Float64})
    for i ∈ 1:L
        for n ∈ 1:N
            acc[n] += n
        end
    end
    sum(acc) # no allocs if sum(acc) is removed
end

function accum2(L, ::Val{N}) where {N}
    acc = zero(MVector{N,Float64})
    for i ∈ 1:L
        @inbounds for n ∈ 1:N
            acc[n] += n
        end
    end
    sum(acc)
end

function accum3(L, ::Val{N}) where {N}
    acc = zero(MVector{N,Float64})
    for n ∈ 1:N # switch loop order - slower, but no allocs
        for i ∈ 1:L
            acc[n] += n
        end
    end
    sum(acc)
end

@btime accum1($10,$(Val(4)))   #  21.393 ns (1 allocation: 48 bytes)
@btime accum2($10,$(Val(4)))  #  7.710 ns (0 allocations: 0 bytes)
@btime accum3($10,$(Val(4)))  #  21.731 ns (0 allocations: 0 bytes)

Start by reproducing the allocation measurements for accum1, accum2, and accum3 with the supplied Julia examples, comparing the effects of @inbounds and loop order. Trace the StaticArrays MArray behavior involved in zero, indexing, and sum; done means explaining the inconsistent allocation and establishing a consistent result without changing the intended accumulation behavior.