[clr-interp] Fix dynamic methods

[davidwrighton]

• We needed to clear the interpreter code pointer
• We needed to re-use any InterpreterPrecode that are associated with a dynamic method

[Copilot]

Pull Request Overview

This PR fixes dynamic methods in the CLR interpreter by addressing two key issues: clearing interpreter code pointers during method reset and reusing InterpreterPrecode instances for dynamic methods.

• Adds a ClearInterpreterCodePointer() method to properly reset interpreter state
• Introduces logic to cache and reuse InterpreterPrecode instances for dynamic methods
• Integrates interpreter code clearing into the method reset process

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
src/coreclr/vm/method.hpp	Adds ClearInterpreterCodePointer method and m_interpreterPrecode field to DynamicMethodDesc
src/coreclr/vm/method.cpp	Integrates interpreter code clearing into MethodDesc::Reset()
src/coreclr/vm/jitinterface.cpp	Implements InterpreterPrecode reuse logic for dynamic methods

[dotnet-policy-service]

Tagging subscribers to this area: @mangod9
See info in area-owners.md if you want to be subscribed.

[kg]

Are we properly resetting the precode associated with the dynamic method? Are transitions between 'interpreted dynamic method' and 'jitted dynamic method' possible at runtime, and do we care about them? If I understand the state of this PR correctly once a DM has interpreter precode it will always have interpreter precode, and if it has JIT precode it will always have JIT precode. I don't totally understand what's going on with the Reset stuff here - it looks like from grepping around we maintain a freelist of dynamic methods for reuse (after a dynamic method is GCd?)

[davidwrighton]

@kg Not quite, this change makes it so that if we allocate an InterpreterPrecode for a MethodDesc, that is the one we'll always use, but it doesn't guarantee that we'll stay using an InterpreterPrecode. MethodDesc::Reset puts us back to having a normal entrypoint which can in theory go either way.