math/remainder
===============================================================================
%% Floating-point remainder of division
1. Usage
-------------------------------------------------------------------------------
```mech:disabled
Y := math/remainder(X, Divisor)
```
2. Description
-------------------------------------------------------------------------------
Computes the IEEE 754-style floating-point remainder of the division
X / Divisor. The remainder has the same sign as X and magnitude less
than or equal to half the magnitude of Divisor.
This function supports scalar values, vectors, and matrices in both
single-precision (f32) and double-precision (f64). It applies
element-wise to structured inputs.
3. Input
-------------------------------------------------------------------------------
| Argument | Kind | Description |
|----------|----------------|--------------------------------------------------|
| X | float, [float] | Dividend value(s). |
| Divisor | float, [float] | Divisor value(s). |
4. Output
-------------------------------------------------------------------------------
| Argument | Kind | Description |
|----------|--------------|---------------------------------------------------|
| Y | matches input| The floating-point remainder of X / Divisor. |
5. Examples
-------------------------------------------------------------------------------
(a) Compute remainder of two scalars
```mech:ex1
y := math/remainder(5.3, 2.0)
```
(b) Compute element-wise remainder of two vectors
```mech:ex2
x := [5.3, 6.7, 9.1]
d := [2.0, 2.5, 4.0]
y := math/remainder(x, d)
```
(c) Compute element-wise remainder of two matrices
```mech:ex3
x := [5.3, 6.7; 9.1, 10.5]
d := [2.0, 2.5; 4.0, 3.0]
y := math/remainder(x, d)
```
6. Details
-------------------------------------------------------------------------------
The remainder is calculated using the libm functions `remainder` (f64)
and `remainderf` (f32), which follow the IEEE 754 definition. For scalars,
the result is the remainder of X divided by Divisor.
- For X = NaN or Divisor = 0, the result is NaN.
- For infinite Divisor, the result is X.
- For infinite X and finite Divisor, the result is NaN.
The function is implemented in Mech with macros for scalar, vector, and
matrix cases. Specialized implementations are provided for various
matrix and vector dimensions (Matrix1, Matrix2, …, Vector2, Vector3, …).
It integrates with the Mech runtime by implementing the MechFunctionImpl
trait, enabling evaluation and compilation.
This makes math/remainder suitable for element-wise numeric computation
across scalars, vectors, and matrices in numerical and scientific
applications.