When you stop recording, the last CC7 message will be the position of the controller at the point you stopped. Even if you play a few notes afterwards, before you stop recording, you have stopped moving your volume controller.
For it to last to the end of the project, you have to record or “tell it” what you want the volume to be.
So, for instance, if your midi part ends at 1 minute and your track is 2 minutes, you could create another part (set L & R makers at 1 min & 2 mins) and double click to create an empty midi part. Then you could draw a CC7 curve in - perhaps a fade out.
The question is “why”?. If you have stopped recording notes, why do you need the volume to go to the end of the project?
Just to clarify a possible source of confusion…
Are you entering your volume curve by recording into a MIDI Part (which is how it reads, to me ), or by using Write/Read Automation (which will actually move the MIDI fader in Cubase’s Mixer?
If the former, then I’m afraid Jaslan’s suggestion about “Fill to End” doesn’t apply here.
But that still doesn’t explain why CC#7 is getting cut off at the end of a bar. (Does this happen, wherever in the song you are, or are you speaking of one specific bar?). Is there some other MIDI track, on the same channel, containing a Part which ends at the point where your CC#7 is getting reset?
(Btw, I share Tacman7’s opinion about the use of CC#7 in general … it is good for balancing the various channels of a multi-timbral instrument, but, wherever possible, much better to set the audio volume of the instrument’s return channel(s).