fix: honcho auth + deepseek LLM config

ai/honcho/config.toml

@@ -48,8 +48,8 @@ FLUSH_ENABLED = true
 
 [deriver.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 # --- Dialectic ---
 [dialectic]
@@ -61,36 +61,36 @@ MAX_TOOL_ITERATIONS = 1
 MAX_OUTPUT_TOKENS = 512
 [dialectic.levels.minimal.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 [dialectic.levels.low]
 MAX_TOOL_ITERATIONS = 3
 [dialectic.levels.low.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 [dialectic.levels.medium]
 MAX_TOOL_ITERATIONS = 2
 [dialectic.levels.medium.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 [dialectic.levels.high]
 MAX_TOOL_ITERATIONS = 4
 [dialectic.levels.high.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 [dialectic.levels.max]
 MAX_TOOL_ITERATIONS = 10
 [dialectic.levels.max.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 # --- Summary ---
 [summary]
@@ -100,8 +100,8 @@ MESSAGES_PER_LONG_SUMMARY = 60
 
 [summary.model_config]
 transport = "openai"
-model = "hermes-3"
-base_url = "http://ollama:11434/v1"
+model = "deepseek-v4-flash"
+base_url = "https://opencode.ai/zen/go/v1"
 
 # --- Dream ---
 [dream]