Re-organize data reduction

[ArneBouillon]

This PR makes the following changes:

• The orig2zscore and zscore2orig functions, which were unused, are removed.
• The individual data processors are moved to their own files.
• Decorrelater is renamed to Decorrelator
• Instead of exporting all subroutines related to data processing, I now only export create_encoder_schedule, initialize_and_encode_with_schedule!, encode_with_schedule, and decode_with_schedule. My rationale for this is twofold.
  • It makes clear that these are the APIs that should be used in practice. The internal functions now start with an underscore (can be removed! just an idea) and can, if necessary, still be used outside the package as, e.g., CES.Utilities._initialize_and_encode_data!.
  • It allows the internal subroutines to implement only the methods necessary for our API functions, reducing the number of methods we need to define and maintain.
• Partially thanks to the above change, I could remove the extract_data functions.

I also started working towards LikelihoodInformed processors by passing both structure matrices to the initialization code of PDCPs. I didn't pass the observation yet, since I'm still looking for a clean way to add this to the pipeline.

[odunbar]

Much cleaner!

I'm very glad that we will remove the anonymous functions with these changes. I will just comment now, and approve on failed runtests. These seem almost entirely based on functions that have been removed and are tested.

You may also need to check that there is not entry for the normalizations z2orig... etc. in docs/src/API/Utilities.jl or docs may have issues.

[odunbar]

I have addressed some of your comments in the PR so there may be small conflicts - I will refrain from further changes until this is merged. (and will work on a Docs page)

[codecov]

Codecov Report

Attention: Patch coverage is 97.24771% with 6 lines in your changes missing coverage. Please review.

Project coverage is 91.12%. Comparing base (7ad963c) to head (8d33f8f).
Report is 6 commits behind head on orad/migrate-data-proc-to-new.

Files with missing lines	Patch %	Lines
src/Utilities.jl	95.55%	2 Missing ⚠️
src/Utilities/canonical_correlation.jl	96.87%	2 Missing ⚠️
src/Utilities/decorrelator.jl	96.66%	2 Missing ⚠️

Additional details and impacted files

@@                        Coverage Diff                        @@
##           orad/migrate-data-proc-to-new     #371      +/-   ##
=================================================================
+ Coverage                          91.08%   91.12%   +0.03%     
=================================================================
  Files                                  7       10       +3     
  Lines                               1560     1544      -16     
=================================================================
- Hits                                1421     1407      -14     
+ Misses                               139      137       -2     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[odunbar]

LGTM!