Add documentation to each dataset on download sources, and what files specifically are compatible.

[jhnwu3]

Right now, we just assume the users know where each dataset is located, and exactly what files are required.

But, MIMIC-CXR vs. MIMIC-CXR-jpg are 2 different datasets referencing the same dataset.

MIMIC-CXR-JPG is supported, but MIMIC-CXR is not.

[ParthAgarwalCode]

Hi, I’d be interested in working on this issue if it’s still available. This would be my first contribution to PyHealth, and I’d be glad to take it on.

[jhnwu3]

Go for it.

[sevvaluluss]

I'd like to work on this