Refactor to TS, and jest smoke test

[Beanow]

Besides refactoring to split out the ruleCheck (and TS, oop! my hands slipped there),
this has an example of testing the collected feedback against a known 100% setup.

Since a correct rule should not break when compared to a perfect bias.
One feedback comes up bad currently, you can see this with:

yarn
yarn test --watch

 FAIL  core/core.test.ts
  ✕ Smoke test rule checker (3 ms)

  ● Smoke test rule checker

    expect(received).toEqual(expected) // deep equality

    - Expected  - 1
    + Received  + 9

    - Set {}
    + Set {
    +   Object {
    +     "biasIndex": 3,
    +     "dx": 0.10089285714285717,
    +     "feedback": "good",
    +     "timestamp": 1661958970541,
    +     "value": 0.11517857142857141,
    +   },
    + }

Suggesting (assuming this feedback is correct) the good breakpoint might be slightly off.

export const goodBreakpoint: number = 0.1 + eps;
// versus dx: 0.10089285714285717

---

I understand this isn't exactly a "minimal" PR.
So feel free to pick and choose as you see fit 😄

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Updated
f1-manager-calc	✅ Ready (Inspect)	Visit Preview	Sep 2, 2022 at 5:38PM (UTC)

[iebb]

would test about the 0.1 case, since i haven't precisely test this one than the optimal (39/5600) one. would look into it, but i haven't received feedbacks leading to wrong results because of good/great, so i assume it should be somewhere working.

this code definitely needs lots of rework to meet the quality standards, which ts could be a good entrypoint. would check this carefully when i have time.