ignore_index in loss functions

[erickrf]

This pull request implements two changes to make the loss function API closer to nn.CrossEntropyLoss.

• Treating ignore_index. The common use case is to use a pseudo class id such as -1 in the target tensor to indicate padding positions (or any samples that should be ignored in the loss computation). This PR filters out ignore_index from the target tensor before computing the actual loss; the current implementation does not do this.

• Adding reduction mean as a synonym to elementwise_mean. The latter has been deprecated in nn.CrossEntropyLoss in favor of mean.