fix: reduce sin/cos angle modulo 360 to avoid precision loss

[Chessing234]

Resolves

Resolves #2199

Proposed Changes

operator_mathop's sin and cos cases convert the input angle straight to radians (Math.PI * n / 180). For very large angles this argument accumulates floating-point error, so the result drifts away from the correct periodic value. For example cos(36000000000090) returns 0.000025074 instead of 0, even though the angle is well below Number.MAX_SAFE_INTEGER.

This change reduces the angle modulo 360 before converting to radians. Trigonometric functions are periodic with period 360°, so this is mathematically equivalent for all inputs while keeping the radian argument small and precise. MathUtil.tan already does exactly this (angle = angle % 360), so tan was unaffected — this brings sin/cos in line with it.

Reason for Changes

Since the functions are periodic, they should return the same value for angles that differ by a multiple of 360. Without modular reduction the largest angles silently produce wrong results.

Test Coverage

Added two assertions to the existing mathop unit test verifying that large angles reduce correctly:

t.strictEqual(blocks.mathop({OPERATOR: 'sin', NUM: 36000000000090}), 1);
t.strictEqual(blocks.mathop({OPERATOR: 'cos', NUM: 36000000000090}), 0);

Both fail before the change and pass after. The existing trig assertions (sin 1/90/180, cos 1/180, etc.) are unchanged, confirming no regression for normal angles. npm run test:unit -- test/unit/blocks_operators.js passes and ESLint is clean.

[github-actions]

Thank you for your submission, we really appreciate it. Like many open-source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution. You can sign the CLA by just posting a Pull Request Comment same as the below format.

---

I have read the CLA Document and I hereby sign the CLA

---

_{You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot.}