This adds a function `isequal_normalized` to the Unicode stdlib to check whether two strings are canonically equivalent (optionally casefolding and/or stripping combining marks).

Previously, the only way to do this was to call `Unicode.normalize` on the two strings, to construct normalized versions, but this seemed a bit wasteful — the new `isequal_normalized` function calls lower-level functions in utf8proc to accomplish the same task while only allocating 4-codepoint (16-byte) temporary arrays.  It seems to be about 2x faster than calling `normalize` in the expensive case where the strings are equivalent, and is potentially much faster for inequivalent strings for which the loop can break early.  (If we could stack-allocate small arrays it might get faster.)

(In the future, we might also want to add `Unicode.isless_normalized` and `Unicode.cmp_normalized` functions for comparing Unicode strings, but `isequal_normalized` seemed like a good start.)

This adds a function `isequal_normalized` to the Unicode stdlib to check whether two strings are canonically equivalent (optionally casefolding and/or stripping combining marks).

Previously, the only way to do this was to call `Unicode.normalize` on the two strings, to construct normalized versions, but this seemed a bit wasteful — the new `isequal_normalized` function calls lower-level functions in utf8proc to accomplish the same task while only allocating 4-codepoint (16-byte) temporary arrays.  It seems to be about 2x faster than calling `normalize` in the expensive case where the strings are equivalent, and is potentially much faster for inequivalent strings for which the loop can break early.  (If we could stack-allocate small arrays it might get faster.)

(In the future, we might also want to add `Unicode.isless_normalized` and `Unicode.cmp_normalized` functions for comparing Unicode strings, but `isequal_normalized` seemed like a good start.)

Fixes #52408.

(Note that this function was added in Julia 1.8, in #42493.)

In the future it would be good to further optimize this function by
adding a fast path for the common case of strings that are mostly ASCII
characters. Perhaps simply skip ahead to the first byte that doesn't
match before we begin doing decomposition etcetera.

Fixes #52408.

(Note that this function was added in Julia 1.8, in #42493.)

In the future it would be good to further optimize this function by
adding a fast path for the common case of strings that are mostly ASCII
characters. Perhaps simply skip ahead to the first byte that doesn't
match before we begin doing decomposition etcetera.

(cherry picked from commit 3b250c7)