autoreg process estimates closed-form solution of periodograms

Added the 'autoreg' process to estimate the closed-form solution of periodograms, following https://www.biorxiv.org/content/10.1101/2020.09.23.309724v1. This addresses issue #1 (closed).

In my implementation, the basic functionality proposed in issue #1 (closed) is included. Additionally, the process returns a third dataframe as an output, and this dataframe contains the optimal orders (see b7802cfd).

This process is modelled after the 'fft' process, https://git.ecdf.ed.ac.uk/swain-lab/time-series-pipeline/-/blob/1a2924afb9b7c5c22e68cec8a49c42016824fc55/pipeline/ar_grima2020.py and 'autoreg' in https://git.ecdf.ed.ac.uk/swain-lab/time-series-pipeline/-/blob/1a2924afb9b7c5c22e68cec8a49c42016824fc55/pipeline/periodogram.py

This commit was part of an attempt to fix a bug that affected unit tests.

The bug suggests issues with access tokens.

See output: https://git.ecdf.ed.ac.uk/swain-lab/aliby/postprocessor/-/jobs/21237

This commit continues cbe5ed93 in an attempt to fix a bug that affected unit tests. See relevant console output on: https://git.ecdf.ed.ac.uk/swain-lab/aliby/postprocessor/-/jobs/21237

It was found that the issue lay with setting the protection of variables in global CI/CD setting for this repository. In sum, these variables were protected and thus were only available to use if there was a commit on master. This is not a master branch, so the bug arose. The issue was rectified by manually changing the protection of the relevant variables.

The changes successfully eliminated the error described in the commit message. However, another error was raised instead. See commit 950dab5f

The changes successfully eliminated the error described in the commit message, and the whole pipeline passes.

assigned to @s1947236

I had mistakenly created a merge request to go into the wrong repository at https://git.ecdf.ed.ac.uk/swain-lab/python-pipeline/postprocessor/-/merge_requests/2. It should be merged into https://git.ecdf.ed.ac.uk/swain-lab/aliby/postprocessor/-/tree/master NOT https://git.ecdf.ed.ac.uk/swain-lab/python-pipeline/postprocessor/-/tree/master. I think it was because I was referring to an issue in the python-pipeline project -- which has since been moved to the aliby project, and it automatically assumes I want to merge into that project for some reason.

requested review from @amuoz

unassigned @s1947236

approved this merge request

merged

mentioned in commit 8c609ab5