Add suggest alternatives for Out-of-range \x escapes

[chenyukang]

Fixes #148917

seems add two notes seems better.

r? @scottmcm

[scottmcm]

The two-notes approach looks reasonable to me in the same error, but I really don't know anything about this part of the compiler so I'm going to reroll, sorry.
r? diagnostics

[Kivooeo]

I'm not 100% sure about note here, it feels to me that help would be better, because we are trying to help user to fix their incorrect code but I can be wrong about when to use each so correct me here

[chenyukang]

I'm not 100% sure about note here, it feels to me that help would be better, because we are trying to help user to fix their incorrect code but I can be wrong about when to use each so correct me here

I'm neutral for note or help in this case, anyway, I changed it into help.

[Kivooeo]

last one question and we are good to go: is this intended to have span like this for a string with len more than 4?

[chenyukang]

good catch, i think it's better to point to the span of full literal, fixed it.

[Kivooeo]

I feel like supporting strings is a bit painful, like for example

error: out of range hex escape
 --> /home/gh-Kivooeo/test_/src/main.rs:2:11
  |
2 |      dbg!(" this is some kind of string \xa7");
  |            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ must be a character in the range [\x00-\x7f]

error: aborting due to 1 previous error

which is not 100% correct, while I understand that calculating a precise span would be a challenging, would it be reasonable to remove string support for now and make it char only, it also will make implementation easier from what I see

upd: the current nightly already shows correct span

error: out of range hex escape
 --> src/main.rs:2:40
  |
2 |     dbg!(" this is some kind of string \xa7");
  |                                        ^^^^ must be a character in the range [\x00-\x7f]

[Kivooeo]

oh i see, we can't actually remove string support, I originally thought that we can because of this line, but it actually means not much matches!(mode, Mode::Char | Mode::Str)

so... idk, any ideas on this? Maybe we could somehow extend original err_span