gettext: Update to version 0.24

gettext: Update to version 0.24

- Update from version 0.22.5 to 0.24
- Update of rootfile
- Changelog
    0.24
	# Programming languages support:
	  * JavaScript:
	    - xgettext now parses recursive JSX expressions correctly.
	  * Rust:
	    - xgettext now supports Rust.
	    - 'msgfmt -c' now verifies the syntax of translations of Rust format
	      strings.
	    - A new example 'hello-rust' has been added.
	  * C:
	    - A new example 'hello-c-http' has been added, showing the use of
	      GNU gettext in a multithreaded web server.
	  * C++:
	    - A new example 'hello-c++-gnome3' has been added.
	  * Ruby:
	    - A new example 'hello-ruby' has been added.
	# Improvements for maintainers:
	  * When xgettext creates the POT file of a package under Git version control,
	    the 'POT-Creation-Date' in the POT file usually no longer changes
	    gratuitously each time the POT file is regenerated.
	# Caveat maintainers:
	  * Building the po/ directory now requires GNU make on specific platforms:
	    macOS, Solaris, AIX.
    0.23.1
	* Bug fixes:
	  - Building with libxml2 >= 2.12.0 and gcc >= 14 now works.
	  - XML: The value of the xml:lang attribute, inserted by msgfmt, is now
	    more correct.
    0.23
	# Internationalized data formats:
	  * XML:
	    o The escaping of characters such as & < > has been changed:
	      - No escaping is done any more by xgettext, when creating a POT file.
	      - Instead, extra escaping can be requested for the msgfmt pass, when
	        merging into an XML file.
	      - The default value of 'escape' in the <gt:escapeRule> was "yes";
	        now it is "no".
	      This means that existing translations of older POT files may no longer
	      fully apply. As a maintainer of a package that has translatable XML files,
	      you need to regenerate the POT file and pass it on to your translators.
	    o XML schemas for .its and .loc files are now provided.
	    o The value of the xml:lang attribute, inserted by msgfmt, now conforms
	      to W3C standards.
	    o 'msgfmt --xml' accept an option --replace-text, that causes the output
	      to be a mono-lingual XML file instead of a multi-lingual XML file.
	    o xgettext and 'msgfmt --xml' now support DocBook XML files.
	  * Desktop: xgettext now produces POT files with correct line numbers.
	# Programming languages support:
	  * Python:
	    o xgettext now assumes source code for Python 3 rather than Python 2.
	      This affects the interpretation of escape sequences in string literals.
	    o xgettext now recognizes the f-string syntax.
	  * Scheme:
	    o xgettext now supports the option '-L Guile' as an alternative to
	      '-L Scheme'.  They are nearly equivalent.  They differ in the
	      interpretation of escape sequences in string literals: While
	      'xgettext -L Scheme' assumes the R6RS and R7RS syntax of string literals,
	      'xgettext -L Guile' assumes the syntax of string literals understood by
	      Guile 2.x and 3.0 (without command-line option '--r6rs' or '--r7rs', and
	      before a '#!r6rs' directive is seen).
	    o xgettext now recognizes comments of the form '#; <expression>'.
	  * Java: xgettext now has an improved recognition of format strings when the
	    String.formatted method is used.
	  * JavaScript:
	    o xgettext now parses template literals inside JSX correctly.
	    o xgettext has a new option --tag that customizes the behaviour of tagged
	      template literals.
	  * C#:
	    o The build system and tools now also support 'dotnet' (.NET) as C#
	      implementation.  In order to declare a preference for 'dotnet' over
	      'mono', you can use the configure option '--enable-csharp=dotnet'.
	    o xgettext now recognizes strings with embedded expressions (a.k.a.
	      interpolated strings).
	  * awk: xgettext now recognizes string concatenation by juxtaposition.
	  * Smalltalk: xgettext now recognizes the string concatenation operator ','.
	  * Vala: xgettext now has an improved recognition of format strings when the
	    string.printf method is used.
	  * Glade: xgettext has improved support for GtkBuilder 4.
	  * Tcl: With the recently released Tcl 9.0, characters outside the Unicode BMP
	    in Tcl message catalogs (.msg files) will work regardless of the locale's
	    encoding.
	  * Perl:
	    o xgettext now reports warnings instead of fatal errors.
	    o xgettext now recognizes strings with embedded expressions (a.k.a.
	      interpolated strings).
	  * PHP:
	    o xgettext now recognizes strings with embedded expressions.
	    o xgettext now scans Heredoc and Nowdoc strings correctly.
	    o xgettext now regards the format string directives %E, %F, %g, %G, %h, %H
	      as valid.
	# Runtime behaviour:
	  * In the C.UTF-8 locale, like in the C locale, the *gettext() functions
	    now return the msgid untranslated. This is relevant for GNU systems,
	    Linux with musl libc, FreeBSD, NetBSD, OpenBSD, Cygwin, and Android.
	# Documentation:
	  * The section "Preparing Strings" now gives more advice how to deal with
	    string concatenation and strings with embedded expressions.
	# xgettext:
	  * Most of the diagnostics emitted by xgettext are now labelled as
	    "warning" or "error".
	# msgmerge:
	  * The option '--sorted-output' is now deprecated.
	# libgettextpo library:
	  * This library is now multithread-safe.
	  * The function 'po_message_set_format' now supports resetting a format string
	    mark.

Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>