t-247: Add Provider abstraction for multi-backend LLM support

- Create Omni/Agent/Provider.hs with unified Provider interface
- Support OpenRouter (cloud), Ollama (local), Amp (subprocess stub)
- Add runAgentWithProvider to Engine.hs for Provider-based execution
- Add EngineType to Core.hs (EngineOpenRouter, EngineOllama, EngineAmp)
- Add --engine flag to 'jr work' command
- Worker.hs dispatches to appropriate provider based on engine type

Usage:
  jr work <task-id>                    # OpenRouter (default)
  jr work <task-id> --engine=ollama    # Local Ollama
  jr work <task-id> --engine=amp       # Amp CLI (stub)

1 files changed, 9 insertions, 2 deletions

diff --git a/Omni/Agent/Worker.hs b/Omni/Agent/Worker.hs
index 66f894d..3b0c563 100644
--- a/Omni/Agent/Worker.hs
+++ b/Omni/Agent/Worker.hs
@@ -21,6 +21,7 @@ import qualified Data.Time
 import qualified Omni.Agent.Core as Core
 import qualified Omni.Agent.Engine as Engine
 import qualified Omni.Agent.Log as AgentLog
+import qualified Omni.Agent.Provider as Provider
 import qualified Omni.Agent.Tools as Tools
 import qualified Omni.Fact as Fact
 import qualified Omni.Task.Core as TaskCore
@@ -357,8 +358,14 @@ runWithEngine worker repo task = do
                 Engine.agentGuardrails = guardrails
               }
 
-      -- Run the agent
-      result <- Engine.runAgent engineCfg agentCfg userPrompt
+      -- Run the agent with appropriate provider
+      result <- case Core.workerEngine worker of
+        Core.EngineOpenRouter -> Engine.runAgent engineCfg agentCfg userPrompt
+        Core.EngineOllama -> do
+          ollamaModel <- fromMaybe "llama3.1:8b" </ Env.lookupEnv "OLLAMA_MODEL"
+          let provider = Provider.defaultOllama (Text.pack ollamaModel)
+          Engine.runAgentWithProvider engineCfg provider agentCfg userPrompt
+        Core.EngineAmp -> pure (Left "Amp engine not yet implemented")
       totalCost <- readIORef totalCostRef
 
       case result of