Each class test case generates a basic graph various combinations of custom settings (cycling through each of the possible customizations). The generated graph is then compared to the expected output. Since the output is static, this is easily supported.
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected