`PAR` model handling sample states with missing values

[sarahmish]

Environment details

If you are already running DeepEcho, please indicate the following details about the environment in which you are running it:

• DeepEcho version: 0.2.0
• Python version: 3.7

Question

In PAR model _sample_state https://github.com/sdv-dev/DeepEcho/blob/fb039e6944a2d2633663da7ac881dcc8a7cade1a/deepecho/models/par.py#L470-L472 Sampling from the Bernoulli distribution can yield a possibility of predicting the value as missing, which we then adjust mu to become zero to handle. This will have an effect on the returned data in _tensor_to_data https://github.com/sdv-dev/DeepEcho/blob/fb039e6944a2d2633663da7ac881dcc8a7cade1a/deepecho/models/par.py#L428-L431 This would potentially make us return props['mu'] value for each state we sampled as missing.

We should probably remove L472 and keep mu as is, then _tensor_to_data will handle the case as needed.

The same would be true to the "count" data type as well.