More work on the stdlib

[littlewhitecloud]

Improve stdlib and fix #376

• [x] Add comment about the math lib functions
• [x] Add more math lib functions
• [ ] Write a tutorial about using functions in mem.jou (maybe not lol)
• [ ] Support more function about process.jou
• [ ] ...

[Akuli]

These functions have surprising gotchas that users should be aware of:

• These functions use the current locale, which is os-dependent, and also some projects call setlocale(LC_ALL, "") or similar while others don't. The setlocale() call could also be hidden inside a library, e.g. GTK calls it. So with these functions, you can write code that works in project X on operating system Y, but does not work in a different project or on a different operating system. Or maybe your code breaks when you add a GUI written in GTK.
• They assume that one character fits in one byte (which in UTF-8 text is only true for ASCII characters).
• When these functions don't work as expected, you only notice it with non-ASCII strings. Programmers tend to test their code only with ASCII strings.

Many other languages (e.g. Python) instead give you information about unicode characters in a way that does not depend on the current locale. I'm not yet ready to decide what's the best approach for Jou.

[littlewhitecloud]

These functions have surprising gotchas that users should be aware of:

• These functions use the current locale, which is os-dependent, and also some projects call setlocale(LC_ALL, "") or similar while others don't. The setlocale() call could also be hidden inside a library, e.g. GTK calls it. So with these functions, you can write code that works in project X on operating system Y, but does not work in a different project or on a different operating system. Or maybe your code breaks when you add a GUI written in GTK.
• They assume that one character fits in one byte (which in UTF-8 text is only true for ASCII characters).
• When these functions don't work as expected, you only notice it with non-ASCII strings. Programmers tend to test their code only with ASCII strings.

Many other languages (e.g. Python) instead give you information about unicode characters in a way that does not depend on the current locale. I'm not yet ready to decide what's the best approach for Jou.

Think jou should not depend on setlocale.