Initial distributions of permanent income and mNrm

[Mv77]

(This revives #922, which was merged accidentally and reverted)

#592 asks for "a plausible calibration of distributions at age of workforce entry of: -Permanent income -Assets". In discussions with @llorracc, we determined the best place to get such a calibration would be the SCF. This PR works on adding calibrations of these two distributions coming ---as directly and transparently as possible--- from the SCF.

An issue is that we clearly do not want to have the full SCF in the toolbox. The solution that I propose here is:

• Keep only a relatively light table with all (most of?) the summary statistics we will possibly need: WealthIncomeStats.csv. It currently contains survey-weighted means and standard deviations at multiple age-year-education combinations. I can easily expand it to contain, e.g., deciles in case one wants to avoid lognormality.
• Link to a public repo that produces the table directly from raw SCF files.
• Have functions that parse the contents of the table to produce the exact parameters that HARK's models need.

Please ensure your pull request adheres to the following guidelines:

• Tests for new functionality/models or Tests to reproduce the bug-fix in code.
• Updated documentation of features that add new functionality.
• Update CHANGELOG.md with major/minor changes.

[Mv77]

@llorracc requested that the repo where the table comes from be able to download all the data to produce the table. I've done this now.

If one clones that repo and runs do_all.R it should download and compute everything from 0.

[Mv77]

This is marked [WIP] only because it relies on #930, so that should ideally be merged in first.