I would start with GPS + Galileo and go from there. You really have to try each combination and see what works best for your area. GPS + Galileo should be the best though. Oh, and 1 second recording.
Please note that Galileo will cost you more battery and in most cases in will make little difference. But as ikim1967 said, you have to try for yourself what works best for you.
Galileo constellation is not complete yet. So it's not certain how it will effect it. I did just some very short testing and both Glonass and Galileo was better than GPS but Glonass was best with Galileo better in GPS soak flower tests. Test yourself now...
I haven't noticed a battery difference with different combination - not one that would be noticeable like the great watch face issue.
Personally I haven't seen the improvement I thought Galileo is meant to bring either. Though that could be because the satellites, I see above me, aren't active yet or back to the 5x being somewhat hard of hearing compared to other devices! I think if the 5X was playing 'Marco Polo' it wouldn't win that often. :-/
I'm half a world away in Denver, CO USA. I use GSP for everything, but my Running activity. In Running I have GPS+GLONAS set. The GPS gets a little funky on some of the streets I run down and I have found GPS+GLONAS seems to help.
The best thing I have found to figure out what is better is, do the same activity around the same time and with similar weather. Do it with each GPS setting and then compare the tracks in Garmin Connect and them MyGPSFiles (http://www.mygpsfiles.com/) with GPX exports. You are looking for consistency within the tracks. Is your elevation the same at kilometer 'x' on all three tracks? Is your speed/pace consistent for the km? If your speed was 'x' for kms 2-5, does that match up with the track or does it bounce around.
After you look at the data, pick the setting that works best. You may have different setting for different activities.
I divided my 10K run into pieces to see how GPS, GPS+GLONASS and GPS + GALILEO will perform. I saved my 1K heat up and cool down segments as different activities. And did 3x3KM runs with different options. It was in a running area of the town stadium but there are no tracks to follow, just earth. Teoretically each lap should be 400 meters and at 3 KM's I should finish at just the opposite side of the point I started. I started almost at the same point in all runs. I ran here with my old Fenix 3 and current Fenix 5X. As I have to make a large turn GPS may shorten the route due to watch turned to inside of a big turn. Usually each lap finishes without reaching to the start of the lap on my all Fenix'es. :)
Also in the latest run I used Samsung Health app on my Samsung Galaxy S8 mobile phone to record the track, too. This phone has support for GPS, Glonass, Beidou and Galileo and uses them at the same time. There were about 11 GPS satellites, 8 Glonass and 4 Galileo satellites when I did their runs. Fenix 5 GPS only gave almost the best track in terms of view. But distance was the most off. It finished almost 100 - 110 m earlier than the finish. GPS + GLONASS Here the track is visually very off in one part of the turn. However, distance doesn't seem so off. Actually it's off by about 20 - 30 meters. Much better than the GPS only version. GPS + GALILEO Track seems more deviated then GPS only but better than with Glonass. However, it nailed laps with almost the least loss in the distance. Finish of each lap was closer to the first starting point. GPS only laps finished sometimes 10 meters earlier than the poing they started. :) Also finish point was closest to where it should be. That was just 10 - 15 meters off.
GPS+GALILEO seems to be the winner here. At the start and finish of my 3K run number of Galileo satellites on the sky is reported by GPS Status on my phone as 4. However, sometimes they are 2 and sometimes they're 6. So their help to accuracy may always be different and sometimes may even break compared to other options. Also my first run was faster than the other and I don't know if it affected the correct tracking.
Galaxy S8 result was maybe the worst. I didn't put the phone into power save mode and screen was on but it finished in a very early point. Also I forgot to test the UltraTrack mode. I wonder about it too.
it seems like there is a slight offset between world maps and both Glonass and Galileo location, but the track from Glonass and Galileo is more accurate. IF confirmed it (glonass or galileo) should be better expecially for running, when you need your relative run distance, GPS only should perform better in map based activities, like trial-running
