I'm trying to figure out the maximum possible density for an earth-sized planet before it would become an earth-sized star.
To try to figure this out, I found a reference star, VB 10, known for being among the smallest stars still fully capable of classifying as a star, and compared its features with those of the sun and earth.
Our sun has a radius of about 700,000 km and a mass of about 2E30 kg, giving it a density of around 1.4 g/cm^3. The earth has a radius of about 6,400 km and a mass of about 6E24 kg, giving it a density of around 5.5 g/cm^3. VB 10 has a radius of about 71,000 km and a mass of about 1.5E29 kg, giving it a density of around 100 g/cm^3.
Considering that VB 10 has a radius over 100 times that of the earth, and is very firmly a star at around 18 times the density, the earth could almost certainly be made at least 18 times more dense without becoming a star. But without more examples of high-density low-volume stars to compare, this is a very hard question to answer. Further, it is possible that VB 10 is very far beyond the threshold of becoming a star, meaning that it could be made, say, 100 times less dense and still be fully capable of carrying out fusion, so this method of prediction has some fairly significant flaws.
What does the math concerning conditions for fusion say?