Add testing support to the LLServer

[raffael0]

This PR adds testing support for the LLServer and tests of the Sequencemanager. It also fixes the "start bug" which caused a lot of problems. The "END" field in sequences is now executed as well.

To do this I had to change quite a lot of the code base to even make it somewhat testable. Most notable is that I had to turn a lot of classes into Singletons which I don't really like but at least it works now

Right now I am only testing the start bug and interpolation. More suggestions are welcome :)

Todo:

• Add an automatic test runner with github actions
• Add more Tests

Tests to add:

• 3 value interpolation
• Actions not in Timestamp order. NOTE This does not work rn but I'm postponing the fix for later as it's rather complex and not really urgent
• Sub Timestamps where total subtimestamp time is larger than next parent time step
• What happens when first timestamp is before start/last after end
• Negative timestamp
• What happens when valve position is below for example zero.
• Do non-utf8 characters crash the sequencemanager?
• Test what happens when you have a timestamp in the END field which is after the end of the sequence

I'm looking forward for some feedback on my testing abstractions and how I do my tests before I commit to writing a lot of them.

[miDeb]

I manually merged the main branch into this branch because the automatic merge caused a conflict (the same variables ended up declared twice)

[Copilot]

Pull Request Overview

Adds testing support and refactors file I/O to enable unit tests for the LLServer’s SequenceManager, fixes the “start” bug, and ensures the “END” field is executed.

• Introduce JSON test data and GoogleTest cases for SequenceManager behavior
• Replace direct filesystem calls with a FileSystemAbstraction singleton for mocking
• Update SequenceManager to use the new abstraction, add interpolation helpers, and correct timing

Reviewed Changes

Copilot reviewed 24 out of 24 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
tests/data/SequenceManagerTestData.h	New JSON fixtures for start/execution, interpolation, swaps, UTF8
tests/SequenceManagerTest.cpp	Unit tests for start-once, interpolation, abort, UTF8 handling
tests/CMakeLists.txt	Setup of GoogleTest and test target
src/utility/FileSystemAbstraction.cpp	New concrete implementation of filesystem abstraction
src/utility/LoopTimer.cpp	Added time field but missing update in wait()
src/utility/JSONMapping.cpp	Switched file I/O to FileSystemAbstraction
src/utility/Config.cpp	Switched config load to FileSystemAbstraction
src/SequenceManager.cpp	Integration of FileSystemAbstraction, interpolation refactor
include/utility/LoopTimer.hpp	Made getters const
include/utility/Singleton.h	Added SetInstance() for DI
include/SequenceManager.h	Added interpolation API and new member for filesystem
include/LLInterface.h	Made methods (and destructor) virtual for mocking
include/EventManager.h	Made methods (and destructor) virtual for mocking
CMakeLists.txt	Split into a library and executable; enable testing
.github/workflows/gtest.yml	CI workflow for building and running GoogleTests

Comments suppressed due to low confidence (4)

include/SequenceManager.h:70

• The doc comment is missing the word “time” after “end”. E.g. “The first element is the end time, and the second element is …”.

		 *                           The first element is the end ,

include/LLInterface.h:50

• Declared a virtual destructor but no definition is provided. Add LLInterface::~LLInterface() = default; in a .cpp to avoid linker errors.

		virtual ~LLInterface();

include/EventManager.h:75

• Declared a virtual destructor but no definition is provided. Implement EventManager::~EventManager() = default; or provide an empty definition in the .cpp.

    virtual ~EventManager();

tests/SequenceManagerTest.cpp:8

• [nitpick] This include of <utility> isn’t used in the test. You can remove it to keep dependencies minimal.

#include <utility>

[raffael0]

@miDeb
Can you review the testing abstractions/stuff i moved to singletons?
I think it would be best to merge it as is and then continue in another PR,
since you are also basing your PR on it