1. Fleet interception. Obviously if you see a fleet coming in at a certain vector, its important to shoot them down before they reach your planet or colony. This is however where its most tempting to use unmanned craft like drones, missile buses, and laser focusing elements since you already have targeting data.
2. chasing down stealth craft. I'm sure we all know by now that stealth in space is possible if you know how to cool your ship to look like a random piece of space junk or fit into the spectra of the sun (and other such tricks), but as soon as a stealth craft has finished its mission, it may need to get the hell out of dodge before it gets shot down, and that could mean ditching stealth and using faster, hotter engines. Like submarines, though, the enemy may want to chase it down and capture it intact, which gives you your chase scene.
Oh, and stealth craft are the one kind of manned vehicle that may be tempting to launch in an interception mission, as described before. The other weapons you launch might only damage the fleet, but the stealth craft can continue harassing the invasion fleet for longer because they can't easily be seen. There are even designs out there for stealth fighters and motherships, so just finding one stealth craft doesn't mean there aren't more out there. You can see how this starts to resemble the use of submarines in WWII.
3. Stealth ships encountering each other by sheer bad luck in the depths of space. Every stealth tactic I've seen still has a range limit, and underneath that limit you become detectable again. It might be that both parties to war had the same idea and thus launched their stealth craft on opposing, but identical vectors. Its unlikely, but could still happen, and would be incredibly tense. Each commander would have to decide whether its more important to achieve his own mission, or stop the other ship from achieving their goal, or alternatively, weigh the risk that the other ship is communicating with command via a laser beam message that could compromise him anyway. Shooting could compromise both missions, especially since it would be at incredibly short range (by space standards). Lots of story potential here.
4. Destruction of deep space sensor platforms. I know the objection to stealth in space is contingent on there being sensor platforms behind enemy lines (a feat which itself requires stealth), but that's incentive for both sides to find and shoot down these sensor platforms so that they can launch stealth ships or even entire fleets under the cover of stealth umbrellas. This way you don't have to worry about your enemy launching an interception mission that damages or destroys your invasion fleet.
5. Space stations and space colonies located at Lagrange points. Earth has at least one large Trojan asteroid which would be tempting to hollow out and colonize, and Jupiter has tons more in its Lagrange points which I think have more ice trapped in them than the main belt asteroids tend to have.
Now having said that, I feel I must address Eternal_Freedom here:
This is not relevant, because the "geography" everyone else is concerned with is that which you can literally hide behind so the enemy can't see or shoot you. Islands and the like. In open ocean, the dynamics of the fight are very different than on the coast or in an archipelago. The same is true in deep space. So while the gravitational topography of space means there is a "reason" for placing stations in Lagrange points, it doesn't change the fact that a battle there only has the station itself to use for cover (which may be a warcrime if its a civilian installation like an O'Neil cylinder), which is vastly less cover than a planet or moon can provide in orbital combat. Combat and chases in deep space are much more predictable than combat near planets for this reason, unless stealth technology is involved.Again though, those still have "terrain" of a sort - think of it as "anything that constrains the battle area" rather than geographic features.
And I guess that's one reason I always bring up stealth technology in these discussions. Not only could it actually be real, its a very good storytelling tool for getting around the things that make space combat potentially boring. It opens up another aspect of the fight: intelligence and information management. Maybe the fight hinges on news about whether or not a certain sensor platform is still operational, or having a fleet commander risk spreading out so they can detect and locate a stealth craft via multiple solar occultation events. Now your characters have to think their way to victory, not just pray that they win by sheer numbers, luck and firepower.
6. Statites and quazi-orbital stations. Similar to the above, but using solar sail technology to make a spacecraft orbit more slowly than its orbital distance would usually allow (a "quazite"), or even allowing a spacecraft to stay completely stationary relative to the sun (a "statite"). Unlike stations in Lagrange points, these ships can be placed at any arbitrary point in solar orbit, justifying encounters with them anywhere in deep space you would like. However, the solar sails are likely going to be so big that everyone in the solar system can find them with trivial ease (though statites are basically invisible at interstellar distances to planet hunting telescopes. A quazite on the other hand would be a dead giveaway for intelligent life in the system, as their orbit is too slow for their orbital distance). A tempting place to put a quazite would be between the Earth and the Sun, with an orbit slowed to match Earth's orbit, thus allowing it to detect solar flare events before they can hit the Earth. Just as one example of why you might make such a thing.
That's all the reasons I can think of at the moment why enemy ships might encounter eachother in empty space, but I'm sure there are more.
