1. delete left
2. insert right
In UNIX terms you can just spec out "cp mine yours" as the edit script, which is O(1) from the perspective of the diff algorithm. Applying that script is O(N), of course, but generating it is constant.
1: Super pedants will note that O(N + M) isn't terribly meaningful here since adding two linear terms means you can reduce that to O(N) under the big-O rules. However, I've always like keeping the information like that in the O notation, in some sense precisely because O(N+M) is-equivalent-to O(N) in big-O terms, but if you have additional knowledge like for some reason, M is exponential in a term you care about you can generate the new big-O term that you care about. In graph theory, my professor tended to want to express O(edges) as O(nodes^2), but I always liked retaining the O(edges) because O(edges) yields useful information if you know you have a sparse graph, for instance, whereas O(nodes^2) removes that. To his credit, he never marked me down for it, which puts him ahead of a lot of professors and their personal preferences...
By contrast, graphs can have significantly different relationships between nodes and edges, for instance, with sparse vs. dense graphs.
One could construct a situation in which file size inputs are varying related to each other in some O(...)-significant way, but it would be very strained, especially in the context of a diff algorithm (why are you trying to diff things significantly different in size at all?). You can fiddle with almost any O(...) pronouncement that way, e.g., "adding two integers isn't O(1)". True enough, but generally, more detail then we are looking for.
There is valuable information in running time being linear in the size of both files. You could define n as the combined size if you wanted to, but that could be confusing.