feat: autonomous self-modifying agent with multi-backend LLM support

Adds SelfModifyLoop — an edit→validate→test→commit cycle that can read
its own failure reports, diagnose root causes, and restart autonomously.

Key capabilities:
- Multi-backend LLM: Anthropic Claude API, Ollama, or auto-detect
- Syntax validation via compile() before writing to disk
- Autonomous self-correction loop with configurable max cycles
- XML-based output format to avoid triple-quote delimiter conflicts
- Branch creation skipped by default to prevent container restarts
- CLI: self-modify run "instruction" --backend auto --autonomous
- 939 tests passing, 30 skipped

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

pyproject.toml

@@ -76,6 +76,7 @@ creative = [
 timmy = "timmy.cli:main"
 timmy-serve = "timmy_serve.cli:main"
 self-tdd = "self_tdd.watchdog:main"
+self-modify = "self_modify.cli:main"
 
 [tool.hatch.build.targets.wheel]
 sources = {"src" = ""}
@@ -97,6 +98,7 @@ include = [
     "src/creative",
     "src/agent_core",
     "src/lightning",
+    "src/self_modify",
 ]
 
 [tool.pytest.ini_options]