'Transpose up half a step' using shortcut OPT-T with a selection beginning with the first bar of line causes the wrong part of the score to be transposed.

This is a weird effect !! 🤥.

Tests carried out.
Note: Define OPT-T to be the short cut for 'Transpose up half a step' in Preferences.
Note: After each test I do CMD-Z to revert to the initial state (hopefully!). —At any rate I do see the MuseScore icon change back to dark, which is a good sign. BUT, is the score REALLY back to its initial state?? Anyway, to keep reloading the score would be much more trouble ...
• I do keep a master copy just in case the score is not in its original state after testing.

(1) Condition:
The test score is in Dmin. Make a selection of the first bar of a line. e.g., in the attached score, bar 6.
Do OPT-T.

Expected result:
Bar 6 is now in D#min, followed by a return back to Dmin from bar 7 to the end.
The notes and chords of bar 6 only are transposed.

Actual result:
The key signature changes to D#min from bar 6–16 !!
There follows a return to Dmin for bars 17–21.
Then the return to Dmin is announced AGAIN at bar 22.
* The notes and chords of bar 6 only are transposed (CORRECT).

(2) Select bars 6–22:
The key signature changes to D#min from bar 6–35 (the end) !!

(3) Select bar 2–35:
The key signature changes to D#min from bar 17–35 !!

Or SOMETIMES from bar 1–35.

(4) Select bar 2 only:
There is NO CHANGE in key signature.

(5) Select bar 17 only:
The key signature changes to D#min from bar 17–21, and then Dmin from 22–35.

NOTE: Bar 17 is now the second bar of line 4, and remains so after the CMD-Z (!).

(6) Select the second bar of line 4 (= bar 17 now).
Pressing OPT-T still causes the transposition bug (let's call it the TB).

Now let's do it with bar 16 selected— Again the TB.

(7)
Now go back to line 1 and select bar 2: No TB.
Same with selecting the second bar of all the other lines, viz., 7, 11, 17, 23, 28, 32: No TB.
* Even for bar 17 this time !!

Generalisation: (a) A selection of any length that begins with the first bar of a line will cause the TB.
(b) A selection of any length that does NOT begin with the first bar of a line will NOT cause the TB. (mostly …)

(8) Insert a RETURN after bar 23. Now bar 24 will be the first bar of line 6.
Result:
(a) A selection of bar 24 causes the TB.
(b) Bar 25— no TB.

(8)
Select bar 11.
Press OPT-T OPT-T

Result:
The key signature changes to Emin for bars 11–14. Then back to Dmin to the end.

(9)
Select bar 11.
Press OPT-T OPT-T OPT-T

Result:
The key signature changes to Emin for bars 11–16. Then back to Dmin to the end.

(10)
I pressed OPT-T without selecting anything.
Expected result: The whole score is transposed up one semitone, wth appropriate key signature changes.

Actual result: The program crashed.

(11)
I pressed OPT-T after selecting a bar number. —Or any object at all ...

Result: No change.

Conclusion:
Well well, it seems to go on forever ... There is something weird about this Bar 22. Is it Love Potion No. 22??
Anyway that should be enough food for thought for the tester ...

MacOS 13.1
MacBook M1 Pro.
MuseScore v OS: macOS 13.1, Arch.: x86_64, MuseScore version (64-bit): 4.0.1-230121751, revision: 9b70a8c

Just a question on procedure. As I understand it, this forum is where a MuseScore bug is first posted. Then after allowance for feedback and discussion, if it seems that it is a new bug, it is formally posted on Github. On the other hand, if the bug has already been reported in Github by someone else, there is no need to go any further. Is that the procedure?
Thanks