Add Phase 6: write tools, shell, and edit tools with reasoning-only fix

Implement 6 new agent tools — write_file, make_dir, delete_file,
str_replace, patch_apply, run_command — bringing the agent from
read-only observer to active code modifier. All write/shell operations
are gated through the existing permissions service.

Also fix a bug where qwen3.5 thinking mode produces reasoning tokens
but no content after tool results, causing the agent to silently exit.
The loop now detects reasoning-only responses, retries twice, then
injects a nudge message to break the model out of its thinking loop.

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

tests/unit/test_agent_loop.py

@@ -60,6 +60,7 @@ def client() -> MagicMock:
 def handler() -> MagicMock:
     mock = MagicMock(spec=StreamHandler)
     mock.usage = None
+    mock.had_reasoning_only = False
     mock.reset = MagicMock()
     return mock