Faster iterate method for TrivialInterlacer

[jishnub]

On master,

julia> o = Ones{Int}(ℵ₀)
ℵ₀-element Ones{Int64, 1, Tuple{InfiniteArrays.OneToInf{Int64}}} with indices OneToInf()

julia> B = ApproxFunBase.BlockInterlacer((o, o))
ApproxFunBase.BlockInterlacer((Ones(ℵ₀), Ones(ℵ₀)))

julia> @btime first($B, 10);
  1.212 μs (28 allocations: 2.31 KiB)

julia> @btime Fun($(x->x^2), Chebyshev(-1..0)+Chebyshev(0..1));
  16.448 μs (140 allocations: 9.77 KiB)

This PR

julia> @btime first($B, 10);
  38.610 ns (1 allocation: 224 bytes)

julia> @btime Fun(x->x^2, Chebyshev(-1..0)+Chebyshev(0..1));
  14.697 μs (72 allocations: 4.97 KiB)

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: -0.58% ⚠️

Comparison is base (4f33475) 71.88% compared to head (002687e) 71.31%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #503      +/-   ##
==========================================
- Coverage   71.88%   71.31%   -0.58%     
==========================================
  Files          81       81              
  Lines        8546     8547       +1     
==========================================
- Hits         6143     6095      -48     
- Misses       2403     2452      +49     

Files Changed	Coverage Δ
src/LinearAlgebra/helper.jl	68.77% <100.00%> (-0.15%)	⬇️
src/Operators/banded/Conversion.jl	87.09% <100.00%> (+0.65%)	⬆️
src/eigen.jl	96.42% <100.00%> (ø)

... and 15 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.