I was working with a student of mine on a paper, and we completed about %90+ of the work - implementation, experiments and paper writing. He had to graduate before his student visa was up, so after he defended I signed his thesis forms. I had asked him to share the latest version of the code and improve the paper to submit but he basically disappeared and does not answer my emails. I also noticed that he stopped sharing the earlier versions of the code (on Google Drive which I ran on Colab, it seems I did not save copies). He was a student I trusted and of good character so I never thought he would just disappear. Obviously, the way he behaves makes me skeptical, and I intend to repeat experiments and a new student agreed to take the work, even though it will be a great loss of time and effort to implement everything from scratch.
Moreover, it is basically this student's thesis work and I want to submit the paper with his name on it (first author) but I may need his approval or even a copyright declaration form to publish. How can I proceed in this case? How can I salvage the paper in a situation like this?
EDIT:
I am asking what people do in these situations, and how they salvage the work. I am not invalidating his contribution or authorship. This must be pretty common but it is the first time I am facing this.
I have projects where students work together, and our school mandates students to "publish" in order to graduate. Please note that this could potentially be very problematic for student co-authors as it could threaten their graduation.
Go easy with the assumptions, I had weekly meetings with the student to advise him, the idea was mine and I was very involved in paper writing and editing. I just did not do a good job with versioning/keeping copies of the code. I could potentially replicate everything but it does not technically make him less of a co-author as far as I am aware. Also, as the school requested upon graduation we signed a copyright form - legally we are both co-authors of his thesis work (which the paper is derived from).