Fix bug in systematics

[emilydolson]

This pull request:

• Fixes the bug @amlalejini identified where the MRCA does not get updated if it is alive and then dies at a time when it dying should cause the MRCA to change.
• Incorporates @abbywlsn's updates to systematics documentation, tests, and features from her summer WAVES project
• Fixes Update DataFile tests #397 by removing floating point numbers from the file comparison test and making a separate floating point test (I know it's not technically systematics, but it was broken by the last systematics update)

[codecov]

Codecov Report

Merging #419 (b6bbde0) into master (87a9b3c) will increase coverage by 0.40%.
The diff coverage is 98.29%.

@@            Coverage Diff             @@
##           master     #419      +/-   ##
==========================================
+ Coverage   82.71%   83.11%   +0.40%     
==========================================
  Files         297      298       +1     
  Lines       37060    37692     +632     
==========================================
+ Hits        30654    31328     +674     
+ Misses       6406     6364      -42     

Impacted Files	Coverage Δ
include/emp/Evolve/Systematics.hpp	97.73% <97.67%> (+4.35%)	⬆️
tests/Evolve/Systematics.cpp	99.09% <98.15%> (-0.91%)	⬇️
include/emp/Evolve/SystematicsAnalysis.hpp	100.00% <100.00%> (ø)
include/emp/Evolve/World.hpp	96.37% <100.00%> (+0.01%)	⬆️
tests/Evolve/Systematics_phylodiversity.cpp	100.00% <100.00%> (ø)
tests/data/DataFile.cpp	98.50% <100.00%> (+8.50%)	⬆️
... and 3 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 87a9b3c...b6bbde0. Read the comment docs.

[emilydolson]

Okay, this is ready for review. I know the diff looks scary, but that's mostly just because I rearranged Systematics.hpp so that the methods weren't randomly scattered across the file. The only substantive changes in this PR are:

• Adding/improving tests
• Documentation
• Fixing the bug mentioned in the description
• Adding the very start of some systematics normalization tools
• Removing the time-related arguments to AddOrg and RemoveOrg in the systematics-manager. They made it too easy to accidentally call the wrong version of the function and are unnecessary now that systematics managers track time internally.
• Fixing a few places where the systematics manager was still using ints instead of world positions.
• Subtly change the order of events in world::Update() so that the systematics manager is updated before on placement signals are triggered for synchronous worlds.

[emilydolson]

Superseded by #451