C++: Fix cartesian-like join in ExternalFlow.qll

[MathiasVP]

The signatureMatches predicate only related the Function column to the i column (and not to the name or type columns) which meant that it returned waaaayyyy to many Functions.

Functionally the behavior was correct since the resulting Function was properly related to those entities in the caller (i.e, the interpretElement0 predicate). But it did mean we had a rather large blowup in the number of results for this predicate (which our tests also showed).

Thanks for highlithing this to me, @jketema!

Commit-by-commit review recommended.

[Copilot]

Pull Request Overview

This PR refactors the C++ dataflow external flow signature matching logic to properly utilize namespace information when identifying functions. Previously, namespace was noted as "not being used" in a comment, but this change integrates namespace checking throughout the signature matching process.

Key changes:

• Introduces a getFunction helper predicate to centralize function identification logic based on namespace, type, and name
• Extends signatureMatches predicate to include namespace as a parameter, enabling proper namespace-aware signature matching
• Refactors interpretElement0 to use the new helper, improving code maintainability by reducing duplication

Reviewed Changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated no comments.

File	Description
cpp/ql/lib/semmle/code/cpp/dataflow/ExternalFlow.qll	Adds getFunction helper predicate, updates signatureMatches to use namespace parameter, refactors interpretElement0 to leverage new helper, removes outdated comment about namespace not being used, adjusts classHasQualifiedName bindingset annotation
cpp/ql/test/library-tests/dataflow/taint-tests/test_mad-signatures.ql	Updates test predicate arity from /5 to /6 to match the new signatureMatches signature

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[MathiasVP]

Please disregard the "Pull Request Overview" comment by Copilot. It completely misses the point of this PR 😭

[jketema]

Note that the coding standards failure is my fault, please ignore. I know how to fix this.

[jketema]

There seems to be a slowdown in DCA.

[MathiasVP]

There seems to be a slowdown in DCA.

I'll take a look. Note that the majority of the slowdown is coming from the unstable macOS project. But worth double checking anyway!

Edit: Yeah, it looks like the code generated for the new version of signatureMatches isn't as good as the code on main. I'll need to look at that on Monday 😭