run: wire SessionToolFactory + PostRun artifacts + AttachImages

The session-tool TYPES already lived in tool/ (P4 move) but the executor never
used them. This wires them, unblocking artifact-producing host surfaces (mort's
chat API / chatbot / .skill / scaddy) to run on executus:

- run/session.go: steerMailbox (thread-safe message queue) + runSession
  (tool.AgentSession over it: AttachImages → a user-role multimodal message
  injected before the agent's next step) + runPostRun (panic-isolated hook call).
- executor: create the mailbox + set inv.AttachImages BEFORE the toolbox build;
  add inv.ExtraTools + a SessionToolFactory's per-run Tools to the toolbox; defer
  its Cleanup; merge the session mailbox with the critic's nudges into ONE
  WithSteer; after the run, call PostRun with the full transcript
  (runRes.Messages) → Result.PostRunResult (best-effort, never fails the run).
- run.Result += PostRunResult *tool.PostRunResult.
- dropped the now-dead criticBinding.steerOptions (superseded by drainSteer).

Tests: a factory whose PostRun emits an artifact from the output+transcript +
Cleanup lands on Result.PostRunResult; a factory-added tool is callable.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

run/session_test.go

@@ -0,0 +1,94 @@
+package run_test
+
+import (
+	"context"
+	"testing"
+
+	"gitea.stevedudenhoeffer.com/steve/majordomo/llm"
+	"gitea.stevedudenhoeffer.com/steve/majordomo/provider/fake"
+
+	"gitea.stevedudenhoeffer.com/steve/executus/run"
+	"gitea.stevedudenhoeffer.com/steve/executus/tool"
+)
+
+// TestSessionToolFactoryPostRun: a SessionToolFactory's PostRun hook produces an
+// artifact (from the run output + transcript) that lands on Result.PostRunResult,
+// and its Cleanup is deferred.
+func TestSessionToolFactoryPostRun(t *testing.T) {
+	fp := fake.New("fake")
+	fp.Enqueue("m", fake.Reply("hello artifacts"))
+	m, _ := fp.Model("m")
+
+	cleanupCalled := false
+	inv := tool.Invocation{
+		RunID: "r1",
+		SessionToolFactory: func(_ tool.AgentSession) tool.SessionTools {
+			return tool.SessionTools{
+				PostRun: func(_ context.Context, transcript []llm.Message, output string, _ error) *tool.PostRunResult {
+					return &tool.PostRunResult{
+						Artifacts: []tool.Artifact{{Name: "out.txt", MimeType: "text/plain", Data: []byte(output)}},
+						Metadata:  map[string]any{"transcript_len": len(transcript)},
+					}
+				},
+				Cleanup: func() { cleanupCalled = true },
+			}
+		},
+	}
+	ex := run.New(run.Config{
+		Registry: tool.NewRegistry(),
+		Models:   func(ctx context.Context, _ string) (context.Context, llm.Model, error) { return ctx, m, nil },
+	})
+	res := ex.Run(context.Background(), run.RunnableAgent{ModelTier: "m"}, inv, "go")
+	if res.Err != nil {
+		t.Fatalf("run error: %v", res.Err)
+	}
+	if res.PostRunResult == nil {
+		t.Fatal("Result.PostRunResult is nil — PostRun hook not invoked / not attached")
+	}
+	if n := len(res.PostRunResult.Artifacts); n != 1 {
+		t.Fatalf("artifacts = %d, want 1", n)
+	}
+	a := res.PostRunResult.Artifacts[0]
+	if a.Name != "out.txt" || string(a.Data) != "hello artifacts" {
+		t.Errorf("artifact = {%q, %q}", a.Name, string(a.Data))
+	}
+	if tl, _ := res.PostRunResult.Metadata["transcript_len"].(int); tl < 1 {
+		t.Errorf("transcript not passed to PostRun (len=%d)", tl)
+	}
+	if !cleanupCalled {
+		t.Error("Cleanup was not deferred/called")
+	}
+}
+
+// TestSessionToolFactoryAddsTool: tools the factory returns join the run's
+// toolbox and are callable by the model.
+func TestSessionToolFactoryAddsTool(t *testing.T) {
+	fp := fake.New("fake")
+	fp.Enqueue("m",
+		fake.ReplyWith(llm.Response{ToolCalls: []llm.ToolCall{{ID: "c1", Name: "render", Arguments: []byte(`{}`)}}}),
+		fake.Reply("rendered"),
+	)
+	m, _ := fp.Model("m")
+
+	toolCalled := false
+	renderTool := llm.DefineTool("render", "render a preview",
+		func(_ context.Context, _ struct{}) (any, error) { toolCalled = true; return "ok", nil })
+	inv := tool.Invocation{
+		RunID: "r2",
+		SessionToolFactory: func(_ tool.AgentSession) tool.SessionTools {
+			return tool.SessionTools{Tools: []llm.Tool{renderTool}}
+		},
+	}
+	ex := run.New(run.Config{
+		Registry: tool.NewRegistry(),
+		Models:   func(ctx context.Context, _ string) (context.Context, llm.Model, error) { return ctx, m, nil },
+	})
+	res := ex.Run(context.Background(),
+		run.RunnableAgent{ModelTier: "m", MaxIterations: 5}, inv, "go")
+	if res.Err != nil {
+		t.Fatalf("run error: %v", res.Err)
+	}
+	if !toolCalled {
+		t.Error("session-factory tool was not added to the toolbox / not called")
+	}
+}