Correct repo version & doubts about rotation

Thanks to the authors for the very impressive work.
I have worked with your code on the main branch for a while, and I have the following questions:

**1. Which is the updated version of the repo?**
I have noticed later that also the "v2" branch exists, among others, and I figured that it is the more updated version. However, when I tried to run the following script for downloading the data, I incur in an error:

python -m objathor.dataset.download_holodeck_metadata --version 2023_09_23

It seems that this function does not exist. The other scripts of the objathor.dataset module work fine. For context, I tried this with both objathor 0.0.7 and objathor 0.0.8, on the main branch.

Additionally, I have seen many users in the issues referencing to the file "objaverse_holodeck_database.json", which appears to contain data relative to rotation. However, I could not find it in the data I downloaded following your scripts.\

**2. How are the objects aligned?**

I have seen in your paper and in many issues raised that you use GPT4O to annotate all objects so that they have the "front" facing a specific direction, corresponding to 0 degrees of rotation on the Y axis. However, I tried visualizing the 3D meshes of some objects and I found the following misaligment:

I am showing this bed from two different angles. You can see that the axis are not aligned with the object "front".
And yet, generating a scene with this asset creates a scene in which this bed is axis-aligned. I am wondering if I am missing something, as it appears that the model is being rotated in the correct way before generating the scene.

For context, this model has been obtained from 2023_09_23/assets/cbe6e0dcc10a414a84acc5cc08171b87/cbe6e0dcc10a414a84acc5cc08171b87.pkl.gz.

Could you please clarify the above two points?
Thank you