FR: Propagate hyperparameters of optimizers in kedro

[stroblme]

Is your feature request related to a problem? Please describe.
Cannot set learning rate for optimizer(s) individually

Describe the solution you'd like
Introduce individual learning rate for optimizer(s) together with a concept for handling optimizer specific hyperparameters

Describe alternatives you've considered
N/A

Additional context
N/A

[stroblme]

Implemented this issue using an optimizer section in the data_science.yml file. Note that one has to comment out either the combined or split section.

optimizer:
  combined:
    name: Adam #Adam, SGD
    lr: 0.001
  split:
    classical:
      name: SGD #Adam, SGD
      lr: 0.03
    quantum:
      name: SPSA #Adam, SPSA, SGD, NGD, QNG
      lr: 0.04