Strategies for initializing parameters should be configurable

[antimora]

Feature description

Ability to configure the initial parameters for Conv, Linear and Embedding NN ops.

Feature motivation

Currently when Conv2d constructed, the parameters are initialized with random values. For the inference mode, we do not need randomize the weights. In fact, it's wasteful (especially for large models).

Another motivation is that the NN components needs to be flexible and configurable.

(Optional) Suggest a Solution

There should be universal initializer (probably enum) that could be set in NN configurations. The default should be the same init strategy as it's now implemented.