* Generated model base from LightGCN

* wip

* wip example

* add self-connection

* refactor code

* added sanity check

* Changed train batch size in example to 1024

* Updated readme for example folder

* Update Readme

* update docs

* Update block comment

* WIP

* Updated validation metric

* Updated message handling

* Added legacy lightgcn for comparison purposes

* Changed to follow 'a_k = 1/(k+1)',  k instead of i

* Changed early stopping technique to follow NGCF

* remove test_batchsize, early stop verbose to false

* Changed parameters to align with paper and ngcf

* refractor codes

* update docstring

* change param name to 'batch_size'

* Fix paper reference

---------

Co-authored-by: tqtg <tuantq.vnu@gmail.com>
Co-authored-by: Quoc-Tuan Truong <tqtg@users.noreply.github.com>