If a function is invoked with wrong parameter, the invocation should return null (instead of failing the evaluation).

A function invocation returns null if the function is invoked with wrong parameters, or no function exists with this name.

Change the name of some test cases to bring the focus on the invocation of functions.

Add new test cases for a function invocation that return null.

If a function invocation returns null, we want to verify that a failure is reported.

A user of the API can access the evaluation failures to understand why an evaluation returned null. So, we should verify the reported failures.

To verify the failures, we need a new base test class that returns the whole evaluation result, instead of only the return value. Create a new test matcher to increase the readability of the tests and better failure messages.

Migrate the test class from FeelIntegrationTest to FeelEngineTest. The new base test class allows using the EvaluationResultMatchers that can check for reported failures and produces better failure messages.

Replace the internal test methods by using FeelEngineTest and EvaluationResultMatchers.

Align the failure message if no function was found.

Move test case from SuppressedFailuresTest to InterpreterFunctionTest that bundles all function related test cases.

Remove test cases from SuppressedFailuresTest that are already covered in InterpreterFunctionTest.