segregate scheduler module from on-demand regardUpdatedLink for updateOperationKey functionality

[IswaryaaS]

• When the cyclic process for updateOperationKey is in-progress, there is a considerable delay in serving the on-demand regardUpdatedLink for sending UpdateOperationKey when OKM receives regardUpdatedLink from ApplicationLayerTopology on-demand.
• This is observed especially during regardApplication processes. When the cyclic process is in-progress in OKM, regardApplication process fails with "MAXIMUM_WAIT_TIME_TO_RECEIVE_OPERATION_KEY_EXCEEDED".

Proposal: To avoid the above behavior, we can segregate the scheduler module into separate module so that they are not inter dependent on each other.

[PrathibaJee]

closed the PR https://github.com/openBackhaul/OperationKeyManagement/pull/366 without merging

Moving the code as a separate module will not resolve the mentioned problem as NodeJs provides a single threaded runtime environment.

Further, please consider the following approaches ,

1. Notifications received from ALT shall be included in a priority queue. In the cyclic operation , this priority queue shall be considered as priority in every iteration of the for loop and executed immediately.
2. Instead of the forloop , setTimeout shall be used(with exit criteria index > link.length and timeout set to negligible time that doesn’t effect the performance)
3. Worker thread shall be initiated to perform the cyclic operation separately. (Remember , worker threads are more costly , so this shall be considered as last option if there is no other way to solve this problem)