Improvements to function and record inference

[oxisto]

This PR achieves several things:

• We try to infer the return types of function based on their context
• A simple heuristic to better decide whether a function should be inferred in a class or in a global scope
• A simple heuristic to decide whether a record or a namespace should be inferred
• We recursively infer nested namespaces, e.g. A::B::C

[codecov]

Codecov Report

Attention: Patch coverage is 67.30769% with 34 lines in your changes missing coverage. Please review.

Project coverage is 76.24%. Comparing base (d2c7f28) to head (2642a3f).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...fraunhofer/aisec/cpg/passes/inference/Inference.kt	65.51%	1 Missing and 19 partials ⚠️
...raunhofer/aisec/cpg/passes/inference/PassHelper.kt	68.42%	1 Missing and 11 partials ⚠️
...ain/kotlin/de/fraunhofer/aisec/cpg/ScopeManager.kt	0.00%	0 Missing and 1 partial ⚠️
...s/kotlin/de/fraunhofer/aisec/cpg/test/TestUtils.kt	50.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

Files with missing lines	Coverage Δ
.../de/fraunhofer/aisec/cpg/InferenceConfiguration.kt	94.73% <100.00%> (+3.56%)	⬆️
...e/fraunhofer/aisec/cpg/frontends/LanguageTraits.kt	100.00% <ø> (ø)
.../de/fraunhofer/aisec/cpg/frontends/TestLanguage.kt	80.55% <100.00%> (ø)
...de/fraunhofer/aisec/cpg/frontends/cxx/CLanguage.kt	96.00% <ø> (ø)
...ain/kotlin/de/fraunhofer/aisec/cpg/ScopeManager.kt	80.05% <0.00%> (+2.74%)	⬆️
...s/kotlin/de/fraunhofer/aisec/cpg/test/TestUtils.kt	71.87% <50.00%> (+0.29%)	⬆️
...raunhofer/aisec/cpg/passes/inference/PassHelper.kt	80.48% <68.42%> (-0.13%)	⬇️
...fraunhofer/aisec/cpg/passes/inference/Inference.kt	76.28% <65.51%> (-2.47%)	⬇️

... and 5 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[KuechA]

I think the inference rules require some more discussion and a proper documentation to understand later why we did some things and took some assumptions...

[maximiliankaul]

I'm a bit undecided with this PR. I'm approving it, because it overall seems to be doing the job when looking at the tests. I like the positive examples demonstrated in the test cases. However, I'm not 100% sure on all the mechanics involved here. Maybe a short (in house) introduction to the inference logic would be beneficial...
I don't like all the tests being implemented for the CXX language. This inference logic is used for all languages, isn't it? I understand that there are plans of getting rid of the fluent dsl stuff for fronted-independent tests, but having everything only tested with CXX is a big negative point in my view.