MM Pytorch API: TabularTransform for input tabular sequence

[sararb]

Fixes # (issue)

Goals ⚽

• Add support for padding, transforming, and masking sequential inputs data in MM Pytorch backend
• The implemented transform classes should:
  - Support multiple targets
  - Be used for training, evaluation, and inference

Implementation Details 🚧

• Implement TabularBatchPadding to pad a group of sequential inputs
• Implement TabularPredictNext for generating targets of causal next item prediction
• Implement TabularPredictLast for generating targets of last item prediction
• Implement TabularPredictRandom for generating targets of predicting one random item and truncate the sequence so that the random item is at the last position.
• Implement TabularMaskRandom for masked language modeling training (MLM) strategy
• Implement TabularMaskLast for masking last item in the sequence, generally used to evaluate models trained with MLM.

Testing Details 🔍

• Defined tests for padding and the different sequence transformations

[github-actions]

Documentation preview

https://nvidia-merlin.github.io/models/review/pr-1161

[marcromeyn]

Should we rename to TabularPadding?

[marcromeyn]

I think it might be better to break padding out into a smaller PR first, and then do masking afterwards.