CSP_Mid.CBF Subarray configuration for the observing mode PSS Beamforming (Part 2)

Complete implementation  for the observing mode PSS started in PI6.  Add functionality related to Jones Matrices, in the absence of TM and SDP support implement a simulator that generates Jones Matrices so that the scan state machine can be tested. The Jones Matrices generated by the simulator do not provide useful content. Improve test suite (CI), implement more robust error handling. 

Mid.CBF MCS implements functionality required to support scan configuration and execution for observing modes PSS.

Mid.CBF Subarray may be configured to perform:

1. correlation only
2. PSS Beamforming only
3. correlation and PSS beamforming simultaneously in the same sub-array

Note: It is expected that during normal operations, correlation will be always performed simultaneously with PSS beamforming; but the requirements do not require Mid.CBF Sub-array to reject a scan configuration for PSS beam-forming if correlation is not configured. 

Mid.CBF subarray implements  states and modes correctly. 

This feature has the following outcomes:

Story AT5-296:
1. Implemented and tested a Python tool to generate Jones Matrix (JM) json files.
bug fix for make test
added jones matrix generator script
testing commit effect on helm chart updates
2. Updated the TM emulator to provide subscription point for CbfSubarray device to receive JM.

• JM data provide in formatted json file, generated by the script.
  3. Updated CbfSubarray to receive JM data through subscription to TM emulator and forward it to VCCs.
• Subarray receives JM upon change in the TM emulator data.
• JM are able to be updated during the READY and SCANNING observing states.
  4. Updated the VCC to receive JM data from CbfSubarray (for all 4 VCC devices).

Code updates: https://gitlab.com/ska-telescope/mid-cbf-mcs/-/tree/AT5-296-fsp-pss-jones-matrices

Note: it has been observed that per the Mid.CBF DDD, in PSS-BF configuration the VCC implements Jones matrix correction, whereas in the FSP JM are not applied; this required the Jones Matrices to be passed to VCC as well, which represented an increase in scope with respect to the initial plan.

Testing

• VCCs were checked with a test script and using Jive to manually poll the attributes; test depends on https://jira.skatelescope.org/browse/AT5-613 which is ongoing. Tests to be completed to verify reception of correct JM data in VCCs.

The following are significant learning outcomes from this story's development;

• Working within the Kubernetes container orchestration environment to monitor and control the MCS system of devices.
• Setting up Jive/WebJive to interface with the MCS system.
• Ramp-up on Tango device server development, in particular the client-server paradigm and the publish-subscription mechanism.
• Understanding functionality of SKA base classes of devices and current implementation of MCS.

Ramp-up to MCS development involved familiarization with several new technologies and concepts. In addition some development tools were not readily available or had to be changed in order to work (eg. an IDE that fully supports Docker and Kubernetes for debugging, installation instructions for running WebJive).

Story AT5-613: Due to time constrains (especially due to the dependency on SP-1381 and the setup difficulties described in that feature's outcomes) the end-to-end testing (which is within the scope of this story) has been deferred to PI10.