Add streaming completion integration test with mock LLM provider

Adds tests to verify that:
- All streaming chunks are processed before control returns to caller
- Both tool calls in a multi-tool-call stream are executed
- The finished signal properly terminates stream processing

Also adds Agent::new_for_test() to allow injecting mock providers.

crates/g3-core/Cargo.toml

@@ -4,6 +4,9 @@ version = "0.1.0"
 edition = "2021"
 description = "Core engine for G3 AI coding agent"
 
+[features]
+test-support = []
+
 [dependencies]
 g3-providers = { path = "../g3-providers" }
 g3-config = { path = "../g3-config" }