jpdb is also really nice for this (it’s even better for finding out which kanji representation is most often used for a word).
In the top right you can see こんぽん not こんぼん is in the top 6700 most frequently used words, while ねもと is only top 24700.
Also, it shows that ねもと is written 根元 in 87% of cases, and only in 11% of cases it is written 根本.
Which lets you guess that when you encounter 根本 written like this, it’s most likely read こんぽん.
I don’t know how their scraping algorithm works and how they know which reading is intended in written text, so I can say nothing about that, but I trust them 