Each individual cell has a discharge limit. Since all cells are 18650 and nobody will tell us the actual part number or datasheet of the cell they used, this varies around. Another poster has noticed some battery builders are using samsung laptop cells to build bike batteries. Those don't have the high discharge rating of the bike cells. So those batteries are cheap, but not very useful for e-bikes.
But, if you put cells in parallel, then the discharge rating adds up. Theoretically. However, because LiIon batteries act weird when they are in parallel, the BMS has to manage all the stacks to keep them discharging evenly. In that case, you can add up the discharge rate. If you take two LiIon batteries with two different BMS boards, and parallel them, eventually one of them will supply all the load. So in that case the discharge rate of the single battery actually supplying the load rules.
Actually, I've bought two batteries rated at 50 A discharge where only 1 of 14 stacks was working at the BMS board. So 3 A discharge would make the battery voltage collapse to 1/3 the nominal voltage. Got my money back for one of them, the other would work long enough I thought it was the motor overheating until the battery warrenty expired. The acid test that proved I had garbage was a resistor load that tried to draw 3.5 A from the battery, no motor or controller involved.
Your question has some jumbled numbers IMHO. 3000 ma sound reasonable discharge rate for a 4.2 v cell. I don't know where the 10 A came from, unless they are allowing you 4 x 3000 ma in parallel but downrating to 4 x 2500 ma because they know things aren't perfect when paralleling celles. The 50 a batteries I bought had 14 parallel stacks, which is fairly conservative.
If your talking lead-acid batteries, they do parallel with fewer problems than LiIon cells.