I don't want to be offensive, but it seems that the curve25519.java file is a black box for you. Talking about leaking private key bits from those changes is ridiculous.
But if BCNext, Jean-Luc and you refuse to make changes to that file then we just have to live with a (in my opinion annoying) every now and then failing signature/verification system.
Common sense doesn't work in cryptography. The problem is
easily solvable by
another way, so everyone who uses ur fix is just looking for troubles.
The statement that changes in Curve25519 can't lead to private key leakage still require a proof. If u can't provide the math, then u should prove that current implementation was incorrectly translated into Java from the original.