Fixes and tests for apply!

[Krastanov]

A previous pull request fixed a few issues with apply!, but there were still a number of edge cases that were not covered. This addresses such cases, simplifies the checks for whether an embedding is necessary, and properly detects permutations of the subsystem indices.

[Krastanov]

@Abhishek-1Bhatt , could you review this, as it addresses code you worked on?

I also modified some of your tests to use @test_throws instead of a try-catch. The try-catch is not reliable here because one day the methods could have just stopped throwing and the tests would not have caught that.

It does not use the expensive permutesystems anymore, but more importantly, it fixes the following regression from the previous pull request: apply!(state_with_3_subsystems, [1,2], operator_with_2_subsystems) was broken because it was calling the meaningless permutesystems(operator_with_3_subsystems, [1,2]).

[codecov]

Codecov Report

Merging #144 (cea6607) into master (97ecd53) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #144   +/-   ##
=======================================
  Coverage   93.55%   93.55%           
=======================================
  Files          24       24           
  Lines        3056     3059    +3     
=======================================
+ Hits         2859     2862    +3     
  Misses        197      197           

Files Changed	Coverage Δ
src/apply.jl	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[ba2tro]

This looks perfect! Looks like I also missed the fact that embed can permute the system for us given the indices instead of separately calling permutesystems. The issue with the previous version was that we were only calling embed when the basis for state and operator were not equal, but it also needed to be called when its equal but the order of indices isn't ascending.
I see that is_apply_shortcircuit fixes that.