There have been many results in recent years on the natural density of zeta zeroes on the critical line, with the best bound commonly accepted (as far as I'm aware) that
$$\liminf_{t\to\infty} \frac{N_0(t)}{N(t)} > 0.4128\dots$$
by Feng (here $N_0(t)$ is the number of zeroes $\chi$ on the critical line with $|\Im(\chi)|<t$ and $N(t)$ is the number of zeroes in the critical strip where the same condition holds).
This week, a new revision to this paper was currently released, and I stumbled upon it. It claims that
$$\liminf_{t\to\infty} \frac{N_0(t)}{N(t)} = 1.$$
Given that versions of it have been up for the past almost 8 months, I'm imagining it's not correct (otherwise it would have been reviewed and accepted by now), but I don't know enough analytic number theory to refute it and I haven't been able to find any commentary on it. Furthermore, the paper passes many of the "crank tests" (well-formatted, well-written and structured, etc.), so it is unlikely to be complete garbage.
Can anyone with more knowledge of analytic number theory confirm or refute (or find a reference confirming or refuting) this paper's validity?