One way ANCOVA - Data failing assumption of linearity


I'm working on my Msc in Psychology thesis, and I've chosen to explore if there are group differences between Internal vs External self-worth individuals on depression scores using Facebook addiction score as a co variate. However my data has failed the first assumption of linearity. I've subscribed to Laerd statistics so I can kind of understand what I'm supposed to do. But there seems to be no relationship between any of my variables.

Any suggestions??? Urgent help greatly appreciated.