Revise interface for Ising energy level computation

[ccoffrin]

@zmorrell after using the compute_ising_energy_levels function a bit I realized that most often one probability wants these ordered from lowest to highest energy. So I propose to chance the return value from a dict to a list of tuples as done in this PR. Thoughts?

Do like the names of energy and states in the tuple?

[codecov[bot]]

Codecov Report

Merging #20 (4f5673f) into main (37c5ba8) will decrease coverage by 0.01%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main      #20      +/-   ##
==========================================
- Coverage   94.58%   94.57%   -0.02%     
==========================================
  Files           5        5              
  Lines         443      442       -1     
==========================================
- Hits          419      418       -1     
  Misses         24       24              

Impacted Files	Coverage Δ
src/ising.jl	100.00% <100.00%> (ø)

---

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 37c5ba8...4f5673f. Read the comment docs.

[zmorrell]

@ccoffrin I just tried it and this seems good. This will likely make the filtering easier in the Analytics package as well. One thing we may want to consider moving forward is what sort of naming conventions we want to apply between variable names representing states in the form of ising spins, binary vectors, and ints. I think that the convention of spin for ising spins, bin or binary for binary vectors, and state for integer states makes sense. I'll hold off on merging for now to get your take. This naming convention would not require any changes on this branch

[ccoffrin]

Throughout the code now I do belive we are following the convention of spin (vector of -1/1), binary (vector of 0/1) and state (an integer) as you suggest, so I am good with this convention.

So shall we merge this?