bug(windows): notepad in win11 removes extra unicode char when using the ITfRange interface - Text Services Framework

Notepad in win11 removes extra unicode char when using the ITfRange interface - Text Services Framework. `Specfically ITfRange->SetText`

This behaviour has so far only been observed with Notepad on Windows 11 all other complaint TSF apps do not behave this way. I have tested a few other that use the TSF such as lvirus Text Editor.

Steps to reproduce

1. On Windows 11 Install Keyman and a Keyboard.
2. Start Keyman and select the Keyman Keyboard
3. Open Notepad and paste this text `fór` which is U+0066 U+006f U+0301 U+0072. Or a use a Keyboard that allows you to type this text and the output is the decomposed form. Note: Most like euro-latin will out put the normalised characters in the composed form [U+0066 U+00F3 U+0072].
4. Move the cursor to the right of the `ó` and press backspace
5. The `ó` is deleted and the cursor is o the right of the `r`
Expected is the just the diacritic is deleted and the cursor is to the right of the `o`. Leaving `for`