Add abstraction for FRAM ring array

[jeromehue]

Fixes #323

Also, changes PersistentVariables to only take a single Section as template parameter. The three subsections for the redundant storage of the variables are generated inside the class.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 95.09804% with 10 lines in your changes missing coverage. Please review.

Project coverage is 93.63%. Comparing base (c97b363) to head (740db96).

Files with missing lines	Patch %	Lines
Sts1CobcSw/Edu/ProgramStatusHistory.cpp	0.00%	10 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #296      +/-   ##
==========================================
- Coverage   94.08%   93.63%   -0.46%     
==========================================
  Files          36       38       +2     
  Lines         947     1084     +137     
  Branches       15       26      +11     
==========================================
+ Hits          891     1015     +124     
- Misses         56       69      +13     

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

[PatrickKa]

I started with this review at the top of FramRingBuffer.hpp and just added comments while reading along. Only when I was somewhere in Get() I decided to take a step back, look at the bigger picture and try to understand how this ring buffer actually works. As it turns out, it does not do what I thought it does and what I want it to do. I am sorry to say this after taking so long with this review, but please reimplement the FRAM ring buffer to work similarly to an etl::circular_buffer. Of course, we do not need the full API. We only insert elements at the end and never remove anything. We probably also do not need iterators. I think what we need is the following:

• operator[]
• Back()
• maybe Front()
• Size()
• maybe Capacity()
• Push() (just the one inserting a single value)

What I completely forgot so far is that we also need something like an Initialize() function that searches for the current begin and end of the buffer. For this, we must also store a consecutive number along the actual elements. The begin/end is then found by comparing neighboring numbers and checking if the next one is smaller than the previous one.

As a note at the end: Yes, I was too lazy to remove the comments I made so far again, and I didn't take a look at the rest of the code too. The ring buffer implementation must be corrected first.