Normalize move amounts consistently for scrambles

[lgarron]

Some events produce U2' moves at the moment, we should either fix that when building the SearchGenerators (like we used to) or by applying puzzle-specific simplification on the result.

[DougCube]

I've never seen this happen. You should give an example of this to reproduce the issue.

[lgarron]

I've never seen this happen. You should give an example of this to reproduce the issue.

Good idea. This can change at any time for any given puzzle, but it's currently exhibited by:

╭─── /Users/lgarron/Code/git/github.com/cubing/twsearch ───────────────────────────────────────────────────────────────────────────────────────────────────
├─  main  | ✔ 
├─ lgarron@Germain
│ cargo run --release -- scramble 333 | 📋
┴
    Finished `release` profile [optimized] target(s) in 0.06s
     Running `target/release/twsearch scramble 333`
L' D B R F U2 R' B L F U2 L2' U' L2' F2' D' R2' D' F2' D2 B2' U' R2' B2' U'
┬
╰─── ⏱️ 1.223s ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────