Here's the thing: voice-leading rules aren't universal. They are based on the assumptions of a particular style. In particular, the kinds of concerns brought up in the question sound like OP is worried about "chorale style" (SATB?) voice-leading, derived from principles used mostly in 19th century hymnody.
If the soprano leaps up an octave (G4 to G5) and then continues upward to A6, that's simply a pattern that would never occur in 19th-century hymnody. The rules of voice-leading for that style don't apply, as you're no longer writing in that style.
Instead, what should be considered is the reason for trying to maintain a "chorale-like" texture, and what effect breaking various voice-leading "rules" will have. To understand that, one should consider the rationale behind the rules in the first place. So, let's take the options in turn:
1) Overlap S/A
I'm not sure whether you mean a voice overlap (which sometimes means to move one voice past the range of another voice's previous note, in this case skipping the alto up past G4 when the soprano takes G5) or a true voice crossing. If it's a voice overlap, the effect in a large ensemble can be losing track of which line connects with which, and also a feeling of two voices taking larger leaps in the same direction. But voice overlaps of this sort occur all the time in non-chorale style textures, like an instrumental texture with bass voices and two upper voices that have a section where they travel in parallel thirds (including leaps). It's a textural effect, primarily.
On the other hand, true voice crossing (where the alto sings/plays higher than the soprano) can undermine the sense of melody. That can also work in instrumental pieces with proper orchestration, but should be used with caution elsewhere. If voice crossing of the melody is allowed, it's probably best not to make it a "one-off" note that occurs once, which might sound weird. Instead, you might have an inner voice take a sort of "descant role" for several notes or even an entire phrase.
2) Have S and A more than an octave apart
This again is a textural choice. Many instrumental textures frequently will have a solo melodic voice floating high above the other voices for an extended period. The general rule of keeping upper voices within an octave only helps to give a sense of an "ensemble" that are part of the same texture, rather than have a "solo voice" high above the others. It can be workable, depending on the style you're after. But again use caution in having it be a "one-off" event. Switching to a different texture where the top voice becomes more like a solo may be useful for effect for a phrase or section, but having an unusual spacing for a single chord may just sound odd. It's going to be a judgment call.
3) Merge the S and A voices on the note before the octave leap
There's no voice-leading rule against this even in strict chorale style, so there's no reason to avoid having any two adjacent voices come together on a unison for a single pitch. Having two voices together might tend to make that note sound louder, so I suppose that's a potential concern if that's not an effect you want.
In general, the best compositional advice is to avoid something that stands out and potentially sounds odd in context. You generally want to adopt a consistent set of principles for whatever texture/ensemble you're working with, and you want to avoid something that sounds disruptive. Your goal and rationale for incorporating such a leap in the first place may help guide what solution will work best in your particular situation.