Document float encoding in reference

[mdinger]

How floats are encoded needs documentation. I suppose it would go in reference: floating point. If not there, at least somewhere.

[kmcallister]

That section is about the literal syntax specifically.

In the section on machine types we mention:

The machine types are the following:

...

• The IEEE 754-2008 binary32 and binary64 floating-point types: f32 and f64, respectively.

[mdinger]

Add a reference to it then in the literal section. Makes it easier to discover.