fix: None crash in providers, _merge_deepest loop, lock-file false positives, stage() race, cache TTL, config merge, changelog indent, template check

devgen/providers/anthropic.py

@@ -16,4 +16,4 @@ class AnthropicProvider(BaseProvider):
             max_tokens=1024,
             messages=[{"role": "user", "content": prompt}],
         )
-        return message.content[0].text
+        return message.content[0].text or ""

devgen/providers/base.py

@@ -53,7 +53,8 @@ class BaseProvider(ABC):
             )
         chosen_model = model or self.DEFAULT_MODEL
         try:
-            return self._generate(prompt, api_key, chosen_model, **kwargs).strip()
+            text = self._generate(prompt, api_key, chosen_model, **kwargs)
+            return (text or "").strip()
         except Exception as e:
             self._handle_error(e)
 

devgen/providers/gemini.py

@@ -30,7 +30,7 @@ class GeminiProvider(BaseProvider):
                         max_output_tokens=kwargs.get("max_output_tokens", 2048),
                     ),
                 )
-                return response.text
+                return response.text or ""
             except Exception as e:
                 if is_token_limit_error(e):
                     raise  # base class handles with friendly message

devgen/providers/ollama.py

@@ -36,7 +36,7 @@ class OllamaProvider(BaseProvider):
                 f"Ollama returned no choices for model {model!r}. "
                 "The model may not be installed locally."
             )
-        return response.choices[0].message.content
+        return response.choices[0].message.content or ""
 
     def _handle_error(self, error):
         """Override default wrapping — Ollama has no API key to check."""

devgen/providers/openai.py

@@ -17,4 +17,4 @@ class OpenaiProvider(BaseProvider):
             messages=[{"role": "user", "content": prompt}],
             **kwargs,
         )
-        return response.choices[0].message.content
+        return response.choices[0].message.content or ""