[FEATURE] Refactor k-NN parsing in a common place to be able to be reused in Neural plugin

[shatejas]

Is your feature request related to a problem?
Currently adding parameters in K-NN requires changes in Neural. There is a chance of inconsistencies there with this duplication and increases the dev work of making changes.

What solution would you like?
Since opensearch neural plugin takes a direct dependency on k-NN. I would like to see neural getting access to new parameters without code changes

What alternatives have you considered?
Alternative is to maintain the status quo and keep duplicating parameters in query builders

Do you have any additional context?
N/A