Why only mlp2x_gelu pretrained weights when code supports multiple projector types?

I notice that the VideoLLaMA3 codebase has support for multiple projector architectures:

mlp2x_gelu (currently used in all pretrained models)
mlp3x_gelu
linear
simp_spatial_conv
However, all the official released models (VideoLLaMA3-2B, VideoLLaMA3-7B, etc.) only use mlp2x_gelu projector.

Why were other projector types not trained and released as alternatives?
Are there plans to release variants with different projector types?
Can you share insights on the performance comparison between different projector types?

Start with the projector architecture names listed in the issue: mlp2x_gelu, mlp3x_gelu, linear, and simp_spatial_conv. Done would require a maintainer-provided explanation of why released models use mlp2x_gelu, whether alternatives are planned, and how their performance compares; no files or tests are identified.

Written by the indexing model from the issue text.