Cleanup of scripts and replication package

Before we do our final changes, lets cleanup and rearrange our scripts area so it only has the minimal amount of code required for things actually in the paper. We also should (1) rename things so they are easy to follow and (2) ensure there is a clean separation between running the code and generating the figures if we expect these to happen separately.

The biggest change here is that we will remove the FVGQ replication, which might make things a little cleaner. I think we can remove the following at the very least:

[x] https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/dynare_grid
[x] https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/fvgq_replication
[x] https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/rbc_student_t_replication
[x] https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/rbc_sv_replication or https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/rbc_volatility_replication since I am not sure which is the 2nd order one we will keep in the paper.
[x] The _compute and juila01 in the https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts folder

After that, lets rename and reorganize things to be a little bit clearer for the replication, with more coming in the next issue.

Is there any reason we can't just put all of the scripts which do runs in a single folder called run_samplers? That way you could have a simple location where you generate everything into the .experiments or wherever you need it to be, and can call them all together or one at a time. If so,

[x] The https://github.com/HighDimensionalEconLab/HMCExamples.jl/tree/main/scripts/julia_replication folder is the rbc stuff, right? If so then lets rename these files to be have a rbc_ or something like that, which will ensure they are clearly different. Then move them into a main folder.
[x] My proposal would be to rename all of the files which just run the baseline samplers to get the results as something like rbc_run_samplers.sh, which could replace the name of https://github.com/HighDimensionalEconLab/HMCExamples.jl/blob/main/scripts/julia_replication/tables.sh
- Same for the sgu
[x] The model file for the 2nd order stochastic volatility is called rbc_sv_2, right? If so, then we should call that section rbc_sv_replication
[x] The file https://github.com/HighDimensionalEconLab/HMCExamples.jl/blob/main/scripts/julia_replication/tables.sh doesn't actually generate the tables, right? It is just doing the runs that would be kept in the tables? If so, lets rename these to be consistent across all of these. I notice it is called run.sh in the sgu_replication and the rbc_sv_replication.
- My proposal would be to rename all of the files which just run the baseline samplers to get the results as something like rbc_run_samplers.sh, which could replace the name of https://github.com/HighDimensionalEconLab/HMCExamples.jl/blob/main/scripts/julia_replication/tables.sh if you think that makes sense.
- Same with the sgu in https://github.com/HighDimensionalEconLab/HMCExamples.jl/blob/main/scripts/sgu_replication/run.sh which becomes sgu_run_samples.sh etc.
[x] I think it is better to replace the echo text in https://github.com/HighDimensionalEconLab/HMCExamples.jl/blob/main/scripts/julia_replication/tables.sh#L5 with something saying what the run itself is. The tables are generated somewhere else.

Right now things are a little confusing on where the plots and tables are generated. Lets reorganize things a little as well.

[x] Maybe have a folder generate_figures_tables which assumes that everything else has been run (or moved) into the .experiments folder
- That folder would have the special project/manifest in it which is for plotting and not execution.
[ ] Lets standardized on the names for the table and figure generation for the baseline cases.
[ ] Lets also standardize on the basic name for the generation of results from the experiments to be consistent. Right now we have statplots.jl, sumstats_dynare.py and sumstats_julia.py which I think we can rename as sgu_figures.jl and sgu_tables.py etc. to be consistent. Then something a little different for the frequentist stuff and robustness.

HighDimensionalEconLab / HMCExamples.jl

Cleanup of scripts and replication package #137