Summary

This changes append-mode note updates so they stop appending sentence_field.

Problem

The README says append mode should append to media fields:

• Ctrl+m - append to the media fields of the newly added note
• Ctrl+b - append to the media fields of selected notes

However, the current implementation also appends sentence_field.

This causes duplicated sentence content when updating notes created externally, for example by Rikaitan with the new note timer enabled.

Example:

• existing sentence from Rikaitan:

ヤツらの声に<b>現実味</b>が…

• fuller sentence captured by mpvacious:

あの遠さはヤツらの声に現実味が…

• current append-mode result:

ヤツらの声に<b>現実味</b>が…<br>あの遠さはヤツらの声に<b>現実味</b>が…

The existing update_sentence() logic already preserves target-word markup from the stored sentence field. Appending the full sentence field afterward defeats that behavior by concatenating the old and new sentences together.

This also aligns with the existing comment in update_sentence():

• "if the target word was marked by Rikaitan, this function makes sure that the highlighting doesn't get erased."

That comment implies sentence updates should preserve the marked target in the updated sentence, not concatenate both the old and new sentence text.

Fix

Restrict append-mode field joining to:

• audio_field
• image_field
• miscinfo_field
• secondary_field

Leave sentence text handling to update_sentence(), which preserves existing target-word markup without duplicating the sentence contents.

Result

Append mode still appends media, but sentence updates now produce the expected single sentence with preserved highlighting, e.g.:

あの遠さはヤツらの声に<b>現実味</b>が…