Make p-FEM more versatile

[dlfivefifty]

This should work:

julia> P = JacobiWeight(1,1) .* Jacobi(1,1)[:,1:n]
(1-x)^1 * (1+x)^1 on -1..1 .* view(Jacobi(1.0, 1.0), Inclusion(-1.0..1.0 (Chebyshev)), 1:10) with eltype Float64

julia> P'P
(QuasiArrays.QuasiAdjoint{Float64, QuasiArrays.BroadcastQuasiMatrix{Float64, typeof(*), Tuple{JacobiWeight{Int64}, QuasiArrays.SubQuasiArray{Float64, 2, Jacobi{Float64}, Tuple{Inclusion{Float64, ChebyshevInterval{Float64}}, UnitRange{Int64}}, false}}}}) * ((1-x)^1 * (1+x)^1 on -1..1 .* Jacobi(1.0, 1.0)) * (∞×10 BandedMatrix{Int64} with bandwidths (0, 0) with data 1×10 Ones{Int64} with indices OneToInf()×Base.OneTo(10))

Instead we have to do:

julia> P = JacobiWeight(1,1) .* Jacobi(1,1)
(1-x)^1 * (1+x)^1 on -1..1 .* Jacobi(1.0, 1.0)

julia> (P'P)[1:n,1:n]
10×10 BandedMatrix{Float64} with bandwidths (2, 2):
  1.06667    0.0       -0.228571       ⋅             ⋅          ⋅             ⋅             ⋅             ⋅             ⋅ 
  0.0        0.609524   5.55112e-17  -0.203175       ⋅          ⋅             ⋅             ⋅             ⋅             ⋅ 
 -0.228571   0.0        0.457143      0.0          -0.17316     ⋅             ⋅             ⋅             ⋅             ⋅ 
   ⋅        -0.203175   0.0           0.369408      0.0       -0.149184       ⋅             ⋅             ⋅             ⋅ 
   ⋅          ⋅        -0.17316      -1.38778e-17   0.3108     2.77556e-17  -0.130536       ⋅             ⋅             ⋅ 
   ⋅          ⋅          ⋅           -0.149184      0.0        0.268531     -2.77556e-17  -0.115837       ⋅             ⋅ 
   ⋅          ⋅          ⋅             ⋅           -0.130536  -4.16334e-17   0.236501      5.55112e-17  -0.104025       ⋅ 
   ⋅          ⋅          ⋅             ⋅             ⋅        -0.115837      1.38778e-17   0.211352      0.0          -0.0943535
   ⋅          ⋅          ⋅             ⋅             ⋅          ⋅           -0.104025      0.0           0.191066      1.38778e-17
   ⋅          ⋅          ⋅             ⋅             ⋅          ⋅             ⋅           -0.0943535    -1.38778e-17   0.174349
```