- Replace custom Empty message by StatusMessage
- Pre-merge code cleanup

- Eliminated hardcoded attack connection IPs and introduced a more flexible approach.
- Incorporated a new RPC method and protobuf message in the CAD proto to facilitate setting the attack connection IPs by an external component to enable computation of the ML model's performance for testing purposes.
- Introduced a new environment variable in CAD to enable or disable testing of the ML model accuracy.

Rewrote all docstrings to comply with PEP 257 and added docstrings for all functions that did not have it in CAD and AM service implementations.

Added buffer to centralized attack detector component for it to wait until it receives BATCH_SIZE requests until it makes an inference

Removed functionality to not send to AM attack connections that have already been sent once and put DEMO_MODE to false

- Added GetConfiguredACLRules RPC and ACLRules message to AM protobuf file to allow exporting ACL rules created by this component to other components.
- Implemented new RPC in the AM component to allow exporting ACL rules created to other components.
- Increased KPI monitoring aggregation time interval from 5 seconds to 60 seconds.
- Cleaned up CAD protobuf file.
- Added useful logging statements to CAD and AM components.
- Formatted CAD and AM code.
- Removed unused gRPC channels in AM.
- Restructured the cryptomining detector model directory.
- Renamed cryptomining detector model file to include relevant details to help better identify the model.
- Improved cryptomining detector model loading by removing the need for the model filename to be hardcoded in the code.
- Removed old cryptomining detector models.
- Refactored CAD code to improve readability.
- Added script to automatically copy compiled protobuf files to DAD component.
- Updated complete_deploy.sh to conform to the new TFS Release 2 deployment process.
- Updated CAD and AM protbuf file messages and RPCs to improve readability and clarity.
- Updated CAD and AM client and service to use the new RPCs names.
- Changed CAD output message to use the cryptomining detector model filename as identifier.
- Added ConnectionInfo class in CAD component to facilitate comparison and serialization of connection information.
- Removed test statements used to check the time taken by the cryptomining detector to perform the inference using different batch sizes.
- Added a function to properly measure the time taken by the cryptomining detector model to process a batch of connection statistics.
- Added IP addresses corresponding to known cryptomining connections to validate the cryptomining detector model performance in the classification task.
- Implemented several metrics in the CAD component to monitor the performance of the cryptomining detector model in the classification task.
- Implemented a function to export the performance metrics of the cryptomining detector in the classification task to an external file.
- Added script to retrieve performance metrics of the cryptomining detector in the classification task from CAD container.

Modified CAD component to send ML features to DAD

- ML model features have been embedded in the model's own ONNX file instead of being hard-coded in the component and the old model was replaced by this new one.
- A new RPC has been added to the Centralized Attack Detector component that allows other components to know which features the ML model is using.
- Added a short path in the Service component for forwarding service-to-device ACL rules (to be removed in the future).
- Added logging messages on the Service and Device components to test and track ACL rule forwarding (to be removed in the future).

- Added/Updated headers
- Added script to update copyright headers

- Removed unnecessary imports
- Removed unused parameters in create_kpi and commented code inside this function.
- Updated the documentation of the methods according to the input and output.
AM:
- Removed unnecessary imports
- Removed commented function GetMitigration
- Removed old commented configuration of an acl rule.

- Removed unused code from CAD proto
- Added timing analysis code to CAD

- Removed random initialization of unused fields (device_id, endpoint_id, etc.) in the KPI descriptors during KPIs creation as this is no longer required.
- Fixed the logic of the ML confidence L3 cybersecurity KPI.
- Old inference results in the KPIs monitoring buffer is cleared periodically now.
- Added some useful debugging logs in the KPIs monitoring implementation.

Fixed the creation of the multiple cybersecurity KPIs defined for the scenario 3 in the Centralized Attack Detector component.

- Added script to deploy a single component of scenario 3.

- Minor fixes.

Added comments to document each class and method in the Centralized Attack Detector and Attack Mitigator components

Changed grpc calls to client in l3_attackmitigator and l3_centralizedattackdetector components and added attack mitigator port and name to constants

- Enabled the correct code in the Attack Mitigator component to properly configure ACL rules in the corresponding service to block detected attack connections.

Fixed the L3 security status KPIs monitoring in CentralizedAttackDetector component and cleaned the CentralizedAttackDetector service implementation

Fixed L3 security status KPIs monitoring in CentralizedAttackDetector component and added scenario 3 components to my_deploy.sh