The problem is there isn’t a way for Cubase to know what pitch you intended the note to actually be. Only you (or the performer) knows if they wanted to sing an E or an F. From Cubase’s perspective either of those is a valid choice. Even from a musical perspective either note would be acceptable depending on what you are going for harmonically. Determining what pitch is correct is solely at the discretion of the music maker.
Looking at your image based on visual info alone the 1st note is Bb & the 2nd C - with both being a bit sharp. If you line up the events so their top and bottom edges are aligned with the keyboard note shading boundaries they should be “in tune” - but it might not sound right because of the micro-pitches so you still need to adjust by ear. But none of this takes into account that the singer might have been aiming for a Db and missed it by a bunch (e.g. like if I sang it).
I’m puzzled by exactly what kind of line you want. If you line up the top and bottom edges of the events, which are lines, with the boundary of the keyboard shading, which is also a line, then you have placed the event so its average pitch is the same as the keyboard note it is on. How are the lines that currently exist in Variaudio insufficient?