map() for float data type

If a `float` argument is passed to the `map()` function, it truncates them to integer values. This provides erroneous output that may not be immediately noticed. This does not produce a compiler warning.

Please add a `float` variant of `map()`:

```cpp
float mapf(float value, float istart, float istop, float ostart, float ostop) {
return ostart + (ostop - ostart) * ((value - istart) / (istop - istart));
}
```

### Additional context

#### Additional reports

- https://dicks-photon-arduino.blogspot.com/2016/05/40-arduino-c-map-function-vs.html

#### Related

- https://github.com/arduino/Arduino/issues/288

Start by locating the existing map() entry point in ArduinoCore-API and reviewing how it handles numeric inputs and outputs. Done means a float-specific mapping API preserves fractional values for float arguments while leaving the existing map() behavior unchanged.

Written by the indexing model from the issue text.