Soldering to cells is not ideal, if you have to you can, but it will impact the capacity negatively.
Not ideal, why? WILL impact negatively, PROOF?.
I hear this all the time but no one seems to publish research. They assume the heat causes damage and think is dangerous..lions and tigers and bears..oh my. After soldering a couple handfuls, I took a dremel and then cut them open. I could not tell the difference between a new cell and one I soldered, and I tried very hard. I unrolled the entire cell and closely examined the edges from start to finish. And then I started my testing cycles described above. When done correctly, the heat created in similar to what a bLast from a 200-300A spot welder creates, just not as focused. I have the raw data AND I have the video footage to back it up. While I may not be on par with many of the topics in this forum, I am very much on the forefront with DIY 18650 projects. (hence the name, inverter powered by salvaged 18650's). But no need to argue, I have found the 'lab-coat-type's' despise soldering and that's great for the rest of us.
@TerraHertz, if you do not use them, you can sell them for a few bucks, so its a win,win.
edit:*If you are going to build a pack, I can send you a bunch of 18650 cell holders, trays, or whatever else you want/need for the build. I have a shop full of stuff and will be happy to be of service. Shipping is 'no worries', we can do the eBay global thing, its super cheap from here. (That's how I sent the Caddock resistors around the globe and it was just ~$8 a box.)