Reviewing is time-demanding. It makes huge sense to re-review if comments and changes are submitted. While first review takes a lot of time, the second can be very quick.
If authors provide a version that shows the changes - this makes re-review very fast. They also don't have to spend too much time on resonse-to-reviews letter.
Can I simply reject to re-review if such changes-highlighted version is not provided?
You can certainly ask for it but if the journal does not require authors to provide such files, you are not likely to get one. So why ask? Well, by asking you provide the editors with the wish from reviewers to see this as a permanent feature and in the end such changes may be made by the journal in their instructions for authors.
The reason why I think it is unlikely your request will be immediately heard is that editors usually handle many articles and may work under severe time constraints (often on their free time outside of work). Communicating back and forth with authors of varying background requesting new versions of manuscripts is usually a time consuming venture that could delay the process by weeks, depending on the responsiveness of the author.
So, there may be a good reason to make a request but I would not expect the wish to be heard in many cases.
What format for such "changes-highlighted version" would you like? Maybe you could just use a tool like DiffPDF?
Another problem with your original intent is that you're going to re-review by looking only at the "changes highlighted". What if there are other changes, not highlighted by the author?
