Arcsin

Calculates arcsine. Fuzz testing shows that relative error is always smaller than 0.01%. Uses the polynomial approximation functions found in https://dsp.stackexchange.com/a/25771, but chooses between them at x=0.4788 due to differences in the relative errors as can be seen here https://www.desmos.com/calculator/wrfwjhythe

See the desmos link for what functions f and g in the code refer to.

g

function g(int256 _x) internal pure returns (int256)

f

function f(int256 _x) internal pure returns (int256)

arcsin

function arcsin(int256 _x) internal pure returns (int256)

Arcsine function

Name	Type	Description
_x	int256	A integer with 18 fixed decimal points, where the whole part is bounded inside of [-1,1]

Name	Type	Description
[0]	int256	The arcsine, with 18 fixed decimal points