Fix CUDA BatchNorm bugs and add support for NHWC (#19742)
### Description
- Fix incorrect running_mean / running_var in training mode due to
incorrect momentum and missing input mean/var. runnig_var could be
correct, but has a too high epsilon.
- Fix incorrect checks when using NHWC
- Pass NHWC flag to NormalizeDims to get correct new dimensions from
x_shape
- Register missing double operations to get parity between NHWC/NCHW
