Elevating u128-support to default

[ebfull]

I'd like to use u128 only in this library and remove the backup arithmetic, but only if its arithmetic is lowered properly on target platforms that don't support it.

My ideal strategy:

1. Remove feature(i128_type) invocation, since it is no longer needed on nightly or (eventually) the 1.26 beta compiler. Do patch release.
2. Once 1.26 stable is released, change documentation to remove mentions of nightly. CI can be changed to always target stable, except for clippy. (See Change CI to compile on latest nightly #45)
3. When u128 lowering definitely works, remove backup arithmetic and use u128-support arithmetic only. The u128-support feature will do nothing. In the next minor release, remove the u128-support feature.

[scottmcm]

The lowering is definitely incomplete today. Last I was working on it it could lower the operators, but a bunch of platforms didn't have support for passing and returning i128, so that was insufficient. And there's still additional work needed to lower things like i128::checked_add.

[Mark-Simulacrum]

Please don't remove nightly/beta from CI. It'll help make sure we don't regress anything you depend on without knowing it.

[ebfull]

Okay. BTW My CI only tests against nightly when we merge PR's, which isn't that often.