Fix proc-macro panics and wrong context types (#127); clean up macro code (#124)

[Copilot]

The 1.2.0 macro rewrite introduced several bugs causing compile-time proc-macro panics and incorrect context type generation. This PR fixes all of them and improves the macro code structure.

Bugs Fixed

• use_randmut = false panics: resolve_context was called with the full #[mitosis(...)] attribute; ContextArgs::from_meta hit the unknown use_randmut field and .unwrap() panicked. Fixed by ignoring attributes that don't parse as ContextArgs.

• Wrong context type in Mitosis create_context: mitosis_derive passed RandomlyMutable as the trait name to the context helper, generating fields typed as <T as RandomlyMutable>::Context instead of <T as Mitosis>::Context. Fixed by passing Mitosis.

• Empty struct produces invalid syntax: No-field structs left ctx_ident as None, emitting type Context = ;. Fixed by defaulting to ().

• Tuple index suffix: #i where i: usize emitted self.0usize. Fixed using syn::Index::from(i).

Code Improvements

Replaced the old create_context_helper with a resolve_context function backed by ContextKind/ContextInfo types, removing the ambiguous custom_context: bool flag and replacing imperative Vec::push loops with iterator chains.

Example (previously panicked at compile time)

// use_randmut = false took the non-early-return path and panicked
#[derive(Clone, Mitosis)]
#[mitosis(use_randmut = false)]
struct Foo { field: MyField }

// create_context on Mitosis generated RandomlyMutable::Context fields,
// causing type mismatches when Mitosis::Context != RandomlyMutable::Context
#[derive(Clone, Mitosis)]
#[mitosis(create_context(name = FooCtx, derive(Clone, Default)))]
struct Bar { a: TypeA, b: TypeB }

Tests

Added genetic-rs/tests/derive_macros.rs with 19 integration tests covering all four derive macros across named structs, tuple structs, empty structs, create_context, with_context, and the specific regression cases above.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[HyperCodec]

wait why is this merging into main I specified dev