This is simply untrue. People work and live in close proximity to bats throughout much of China and Southeast Asia, including in Yunnan province. The mine workers who got sick in Mojiang in 2012 (where RaTG13 was discovered) were literally cleaning out massive mounds of bat poop.[1] There is research that shows that a non-negligible fraction (up to a few percent) of the population in some areas of Yunnan province have antibodies to novel SARS-related coronaviruses.[2,3] Interestingly, it is not known how the people in these studies were infected, and before they were randomly tested for these studies, they were not aware that they had ever been infected.
> is unlikely because we would have found the intermediate animal by now
There's no reason to expect we'd have found the intermediate species by now. Finding intermediate hosts can be very difficult. For example, it took four decades to identify the likely host species of Ebola, and even so, there's still a huge amount of uncertainty about whether there are multiple host species, and how spillover occurs.[4]
> (3) is unlikely because the first case found was in China
The frozen food hypothesis that the WHO is looking at is that animals that were raised or caught in Yunnan province, slaughtered and frozen, and sent to Wuhan might have been carrying the virus.
Option 4 is unlikely because nobody knew about this virus before it appeared in December 2019. The researchers at the Wuhan Institute of Virology work closely with international scientists (including in the US, Australia, France and Singapore). They regularly publish identifying RNA fragments of the viruses they discover. The viruses that they have isolated and cultured in the lab are well known, because they've published on them extensively, and because they collaborate with international researchers. They have only isolated three SARS-related coronaviruses (the vast majority of the viruses they discover are only detected as RNA fragments, not "live" virus particles), and those viruses are all much more closely related to the original SARS than they are to SARS-CoV-2. The reason for this is that before this pandemic, researchers focused their attention on viruses that were close to the original SARS (such as WIV-1[5,6]). SARS-CoV-2 and its closely related viruses would have been far less interesting to them. The lab leak theory really is a conspiracy theory, because it requires the scientists at WIV to have discovered a virus that they didn't tell anyone about, including their close collaborators abroad, for them to have secretly isolated it, for it to have escaped from a highly secure laboratory, and then for them to have covered it up. You can assert that all these things happened, but there's precisely zero evidence for it.
The alternative is that one of the millions of people who regularly interact with animals that harbor SARS-related coronaviruses got infected, and that as is usually the case, it takes time and painstaking work to determine how, when and where that happened.
1. https://www.nature.com/articles/s41586-020-2951-z
2. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6178078/
3. https://www.sciencedirect.com/science/article/pii/S259005361...
4. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4014719/
This is a generally informative and helpful comment, but I think there's a flaw in this argument. If WIV isolates viruses (from samples), there could be the possibility that a virus escapes before it is isolated and sequenced, or that the virus is not successfully isolated and sequenced but escapes and infects somebody. We do not need to suggest that they had preexisting records of SARS-CoV-2, which I would agree is clearly beyond reasonable speculation.
For the scenario in your second paragraph to play out, the WIV would have to have isolated the virus, then leaked it, but for nothing to published or communicated to collaborators about it for months. This is very unlikely.