3/2 A higher orbit would shorten the percentage of time that the station spends in the Moon's shadow, but it would be orbiting more slowly. Finding the sweet spot with the shortest blackout is a problem for actual astrophysicists, and I've made do with extra batteries, since the best feature of a low polar orbit is that I can easily send a lunar lander to any point on the surface as the Moon rotates beneath the station's orbit. Ain't science great?

2/2 Once they are close enough, the supply craft will decelerate into an elliptical path to intersect with the station 1/2 orbit later (perigee), then decelerate again upon arrival to circularise its orbit with the station. The station is on a "polar" orbit around the Moon, which is difficult to reach, but maximises the time spent in direct sunlight and with line-of-site communication with Earth. There are, of course, still some days every month and year where half the orbit is on the far side of the Moon from the Earth (interrupting comms) and/or the dark side from the Sun (blocking solar panels,) but I believe this is the best option in a 2-body simulation.

1/2 FYI, it does make sense. Ignoring the grey lines (orbits of other satellites,) the cloured lines show a series of manoeuvres that transform the blue orbit (a resupply craft) into the green orbit (the space station). Orbiting clockwise, it goes blue > dotted yellow > dotted purple > dotted green > dotted red > solid green. You can see that each section intersects the next section at a single point, which is where the engine is fired to change the orbit. The dotted purple and yellow orbits are just there to change the axis, while the dotted green line is the first that actually lowers the orbit. Because the station and spacecraft will be at different positions (orange arrows) along the green orbit, the spacecraft will actually have to remain in a slightly wider, slower orbit, so that the station will eventually catch up to it.