Use AlmostBandedMatrix for MIRK Jacobians

[avik-pal]

• Support for MIRK
• Benchmarks
• Release and Register FastAlmostBandedMatrices.jl
• Support in LinearSolve.jl Add support for almost banded matrix LinearSolve.jl#423
• Fix tests (if needed)

[github-actions]

Benchmark Results

	master	fe836c3...	t[master]/t[fe836c3...]
Simple Pendulum/IIP/MultipleShooting(10, Tsit5; grid_coarsening = false)	1.83 ± 0.071 ms	1.86 ± 0.075 ms	0.987
Simple Pendulum/IIP/MultipleShooting(10, Tsit5; grid_coarsening = true)	3.41 ± 0.14 ms	3.41 ± 0.11 ms	0.999
Simple Pendulum/IIP/MultipleShooting(100, Tsit5; grid_coarsening = false)	0.055 ± 0.0039 s	0.0544 ± 0.0039 s	1.01
Simple Pendulum/IIP/MultipleShooting(100, Tsit5; grid_coarsening = true)	1.09 ± 0.011 s	1.12 ± 0.02 s	0.969
Simple Pendulum/IIP/Shooting(Tsit5())	0.178 ± 0.0053 ms	0.179 ± 0.0057 ms	0.991
Simple Pendulum/OOP/MultipleShooting(10, Tsit5; grid_coarsening = false)	4.41 ± 0.29 ms	4.4 ± 0.38 ms	1
Simple Pendulum/OOP/MultipleShooting(10, Tsit5; grid_coarsening = true)	7.87 ± 4.5 ms	7.66 ± 4.6 ms	1.03
Simple Pendulum/OOP/MultipleShooting(100, Tsit5; grid_coarsening = false)	0.139 ± 0.0062 s	0.132 ± 0.0043 s	1.05
Simple Pendulum/OOP/MultipleShooting(100, Tsit5; grid_coarsening = true)	4.82 ± 0.01 s	4.79 ± 0.027 s	1.01
Simple Pendulum/OOP/Shooting(Tsit5())	0.763 ± 0.035 ms	0.709 ± 0.041 ms	1.08
time_to_load	7.36 ± 0.032 s	8.02 ± 0.033 s	0.918

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[ChrisRackauckas]

All of the benchmarks are shooting methods?

[avik-pal]

I realized that after the benchmarks ran 😓

[avik-pal]

[avik-pal]

Needs a patch for underdetermined systems