Allow lossless conversions from float into integral types (#3294)

## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
Some APIs return JSON values that are intended to be represented by
integral types but are returned as floating values e.g `[1.0, -2.0,
25.0]`.

This allows those values to be converted into Integral types.

## Description

This uses a bidirectional conversion to check if a float can be
losslessly converted into a integral type. This can have issues at the
limits of i64::MAX but I think that's probably acceptable. These values
would be represented imprecisely by floats already.

## Testing
Added additional unit tests of the behavior.

## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates

----

_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._