Refactor compiler diagnostics and deprecations

[KorigamiK]

This is my attempt to get sioyek to compile with without errors and warnings from clang.

The changes help fix deprecation warnings while compiling against Qt 6.8, compiler warnings and potential errors.

We go from around 50+ diagnostics down to 1 so thats good. I currently didn't apply any logical changes so we should be fine.
Please let me know about the review

[ahrm]

Thanks, some of these fix genuine bugs (e.g. missing \ in some regular expressions), but I don't care about most of the style warnings. For that reason I will not merge this pr but I will fix the parts where there are actual problems.

[KorigamiK]

Although I have made sure to not include any stylistic changes and only apply changes to potential bugs and deprecation updates, you are feel to apply these changes as needed.
Also I think I made some changes to input.cpp that were unnecessary (adding overrides), so I've fixed them in my last commit which you can check out here KorigamiK@7a5b030

I can create a new pr if needed

[ahrm]

Thanks, I included most of the suggested changes in ed14e38.

This suggestion is wrong though:

            if ((chars_between_last_dot_and_index[i] > 0) && (chars_between_last_dot_and_index[i] < 128) && std::isalpha(chars_between_last_dot_and_index[i])) {
            if ((chars_between_last_dot_and_index[i] > 0) && std::isalpha(chars_between_last_dot_and_index[i])) {

Changing the upper line to the lower line might cause a crash when dealing with non-ascii characters.

[KorigamiK]

chars_between_last_dot_and_index[i] < 128

I think a signed char is always less than 128 so this is always true
(I Might be mistaken here though)

[ahrm]

I think a signed char is always less than 128 so this is always true

Not when working with unicode characters.

[KorigamiK]

I think a signed char is always less than 128 so this is always true

Not when working with unicode characters.

Makes sense, I think the issue might lie in this container definition,

sioyek/pdf_viewer/utils.cpp

Line 544 in ed14e38

std::vector<char> chars_between_last_dot_and_index;

A larger element type may be needed

[ahrm]

Yes, ideally we would use std::vector<int> in that location, but since in that function we only care about ascii characters it doesn't matter.