refactor: remove internal and_then chaining

[Legend-Master]

I noticed this while debugging the compile time in tauri with cargo-llvm-lines

The and_then chaining was introduced in #1384 to record the duplcated custom protocol error

If we would return it in build anyways, let's just check it there instead

[github-actions]

Package Changes Through 2ef05ba

There are 1 changes which include wry with minor

Planned Package Versions

The following package releases are the planned based on the context of changes in this pull request.

package	current	next
wry	0.51.2	0.52.0

Add another change file through the GitHub UI by following this link.

---

Read about change files or the docs at github.com/jbolda/covector

[FabianLars]

Sometimes(?) simple is better :D

[FabianLars]

I'm only on my phone rn so I may be imagining things but if we use hash map.insert with the protocol name as key we will silently over duplicates on the with_protocol method, no?

[Legend-Master]

but if we use hash map.insert with the protocol name as key we will silently over duplicates on the with_protocol method, no?

The newer handler will override the existing one in the builder, but I feel like that's what makes sense to me, since it's a builder that's not being built yet, why not to allow that?

[FabianLars]

I'm okay with that, it's just different than it was with and_then 🤷 also idk if your argument really applies to a builder method you'll often call multiple times without wanting to overwrite the protocol. but yeah, either way works for me, just wanted to make sure it wasn't an accident.

[amrbashir]

Don't merge just yet, I will review it as soon as I can

[Legend-Master]

The reason I put the linux web context check inside the final build is because I think it makes more sense to check it just before the build since it doesn't matter before that point

Anyways, that's fine either ways

[amrbashir]

The reason I put the linux web context check inside the final build is because I think it makes more sense to check it just before the build since it doesn't matter before that point

I moved it there so we can find the logic more easily, it doesn't make sense to delay the check until .build call.