logiguard fork: GPUI xdg-activation keyboard-focus serial fix

Single-commit orphan branch: full zed-industries/zed @ 8c74db0 source tree
with a 3-file patch applied (no upstream history).

Patch (crates/gpui_linux/src/linux/wayland/):
  - serial.rs: add SerialKind::KeyboardEnter
  - client.rs: store wl_keyboard.enter serial; add latest_serial_of()
  - window.rs: activate() uses keyboard-enter serial (Mutter focus gate)

Mutter honors window activation only when the token carries the keyboard-
focus serial from wl_keyboard.enter; GPUI used a stale mouse-press serial.
See docs/tray-window-focus-wayland.md in logiguard.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

script/analyze_highlights.py

@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+"""
+This script analyzes all the highlights.scm files in our embedded languages and extensions.
+It counts the number of unique instances of @{name} and the languages in which they are used.
+
+This is useful to help avoid accidentally introducing new tags when appropriate ones already exist when adding new languages.
+
+Flags:
+-v, --verbose: Include a detailed list of languages for each tag found in the highlights.scm files.
+"""
+
+import argparse
+import re
+from collections import defaultdict
+from pathlib import Path
+from typing import Any
+
+pattern = re.compile(r'@(?!_)[a-zA-Z_.]+')
+
+def parse_arguments():
+    parser = argparse.ArgumentParser(description='Analyze highlights.scm files for unique instances and their languages.')
+    parser.add_argument('-v', '--verbose', action='store_true', help='Include a list of languages for each tag.')
+    return parser.parse_args()
+
+def find_highlight_files(root_dir):
+    for path in Path(root_dir).rglob('highlights.scm'):
+        yield path
+
+def count_instances(files):
+    instances: defaultdict[list[Any], dict[str, Any]] = defaultdict(lambda: {'count': 0, 'languages': set()})
+    for file_path in files:
+        language = file_path.parent.name
+        with open(file_path, "r") as file:
+            text = file.read()
+            matches = pattern.findall(text)
+            for match in matches:
+                instances[match]['count'] += 1
+                instances[match]['languages'].add(language)
+    return instances
+
+def print_instances(instances, verbose=False):
+    for item, details in sorted(instances.items(), key=lambda x: x[0]):
+        languages = ', '.join(sorted(details['languages']))
+        if verbose:
+            print(f"{item} ({details['count']}) - [{languages}]")
+        else:
+            print(f"{item} ({details['count']})")
+
+def main():
+    args = parse_arguments()
+
+    base_dir = Path(__file__).parent.parent
+    core_path = base_dir / 'crates/languages/src'
+    extension_path = base_dir / 'extensions/'
+
+    core_instances = count_instances(find_highlight_files(core_path))
+    extension_instances = count_instances(find_highlight_files(extension_path))
+
+    unique_extension_instances = {k: v for k, v in extension_instances.items() if k not in core_instances}
+
+    print('Shared:\n')
+    print_instances(core_instances, args.verbose)
+
+    if unique_extension_instances:
+        print('\nExtension-only:\n')
+        print_instances(unique_extension_instances, args.verbose)
+
+if __name__ == '__main__':
+    main()