[bazel.build] /rules/language contains outdated information regarding ints, floats, and sets

### Page link:

https://bazel.build/rules/language

### Problem description (include actual vs expected text, if applicable):

This page needs someone to read through and make it up-to-date with https://starlark-lang.org/spec.html.

### Integers

Bazel says:

> Int type is limited to 32-bit signed integers. Overflows will throw an error.

Starlark says:

> Integers may be positive or negative, and arbitrarily large

### Floats

Bazel says:

> The following Python features are not supported:
> - float and set types.

Starlark says:

> The Starlark floating-point data type represents an IEEE 754 double-precision floating-point number. Its [type](https://starlark-lang.org/spec.html#type) is "float".

### Sets

Bazel says

> The following Python features are not supported:
> - float and set types

Starlark says:

> A set is a mutable, iterable collection of unique values - the set's elements. The [type](https://starlark-lang.org/spec.html#type) of a set is "set".

If Bazel does genuinely not support these features, then it should be made obvious that this is a Bazel-specific issue. Otherwise the information should be removed or corrected.

### Where do you see this issue? (include link to specific section of the page, if applicable)

https://bazel.build/rules/language

### Any other information you'd like to share?

No.

Open the linked bazel.build/rules/language page and compare its Integers, Floats, and Sets sections with the linked Starlark specification. Check whether Bazel differs from the specification, then correct or qualify the statements and verify the rendered page.

Written by the indexing model from the issue text.