NUnit2055 should not fire on ClassicAssert

[manfred-brands]

Fixes #953

[Copilot]

Pull Request Overview

This PR fixes issue #953 by preventing the NUnit2055 analyzer (InstanceOf) from firing on ClassicAssert calls. The key changes refactor constraint expression handling to properly distinguish between Assert.That calls (which should be analyzed) and ClassicAssert calls (which should not).

Key Changes:

• Extracted constraint expression evaluation logic into reusable extension methods
• Modified AssertHelper.TryGetActualAndConstraintOperations to handle assertions with omitted constraints (like Assert.That(condition))
• Updated analyzers to use centralized constraint evaluation instead of duplicated logic

Reviewed Changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
ConstraintExpressionPartExtensions.cs	Adds IsTrueOrFalse() method to evaluate if a constraint part represents true/false, considering negation
ConstraintExpressionExtensions.cs	Adds IsTrueOrFalse() method to evaluate constraint expressions, handling empty constraint lists
ConstraintExpression.cs	Makes expression operation nullable and handles null cases for assertions without explicit constraints
InstanceOfAnalyzer.cs	Refactors to use centralized constraint helper and removes duplicate constraint evaluation logic
AssertHelper.cs	Updates to support assertions with omitted constraints and single-argument Assert.That calls
BaseConditionConstraintAnalyzer.cs	Refactors constraint evaluation to use centralized helper, properly handling both modern and classic assertions
InstanceOfCodeFixTests.cs	Adds test case for Is.Not.True constraint pattern
InstanceOfAnalyzerTests.cs	Adds test for Is.Not.False and verifies ClassicAssert is not analyzed

---

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

[mikkelbu]

Looks good. I only had one comment which is a little unrelated to this PR, so I'm also happy to merge this as it is

[mikkelbu]

Should we rather count the number of Nots ? to handle people who have "odd" assert like

var value2 = 42;
Assert.That(value2 > 40, Is.Not.Not.True);

or is this too much work ?

[manfred-brands]

Done. I thought it was far fetched to think about Is.Not.False, but Is.Not.Not.False is simply confusing.

I wish NUnit would not allow .Not.Not and as long as it does we might want to think about a rule to says that .Not.Not is the same as nothing.