The naming of `allow-newer`

By way of context:

* [`Cabal-1.20.0.0`](https://hackage.haskell.org/package/Cabal-1.20.0.0/changelog) (2014-04-20) introduced the `--allow-newer[=pkgs]` option to ignore the upper bound on all dependencies or dependencies on specified packages in Cabal files.

* Stack 0.1.8.0 (2015-11-20) added an `allow-newer` configuration option, to ignore the bounds in Cabal files. It took its name from the Cabal option, despite having a different function.

* [`Cabal-2.0.0.2`](https://hackage.haskell.org/package/Cabal-2.0.0.2/changelog) (2017-07-24) introduced a _scope_ syntax to `--allow-newer`.

* [`Cabal-2.0.0.2`](https://hackage.haskell.org/package/Cabal-2.0.0.2/changelog) (2017-07-24) also introduced an `--allow-older[=pkgs]` option.

* Stack 2.9.3 (2022-12-16) allowed the Cabal files in which bounds would be ignored to be specified (with `allow-newer-deps`, following the form of `allow-newer`).

* Stack 3.1.1 (2024-07-28) allowed `--[no-]allow-deps` on the command line (following the form of `allow-newer`).

This has confused some users, who expect that:
* Stack's `allow-newer` configuration option should function as Cabal's `--allow-newer` option; and
* are not sure if `allow-newer-deps` is `allow-newer-in-deps` (like the _scoping_ in Cabal's option) or `allow-newer-over-deps` (like Cabal's option) - it is the former.

Given the passage of over 10 years, the ship may have sailed, but perhaps more descriptive Stack configuration options would be:
* `ignore-bounds` \ `--ignore-bounds`
* `ignore-bounds-in`

In respect of the bounds in the Cabal files of package versions included in a snapshot, Stack always trusts the snapshot over the Cabal files.