fix(chatd): publish streaming message_part events during compaction (#22410)

## Problem

Context compaction in chatd persisted durable messages for the
`chat_summarized` tool call and result via `publishMessage`, but never
published `message_part` streaming events via `publishMessagePart`. This
meant connected clients had no streaming representation of the
compaction.

The client's `streamState` (built entirely from `message_part` events in
`streamState.ts`) never saw the compaction tool call, so:

- No **"Summarizing..."** running state was shown to the user during
summary generation (which can take up to 90s).
- The durable `message` events arrived after or interleaved with the
`status: waiting` event, causing the tool to appear as "Summarized" with
the chat appearing to just stop.

## Fix

### 1. `CompactionOptions.OnStart` callback (chatloop)

Added an `OnStart` callback to `CompactionOptions`, called in
`maybeCompact` right before `generateCompactionSummary` (the slow LLM
call). This gives `chatd` a hook to publish the tool-call `message_part`
immediately when compaction begins.

### 2. Tool-result streaming part (chatd)

`persistChatContextSummary` now publishes a tool-result `message_part`
before the durable `message` events, so clients transition from
"Summarizing..." to "Summarized" before the status change arrives.

### Event ordering is now:
1. `message_part` (tool call via `OnStart`) — client shows
"Summarizing..."
2. LLM generates summary (up to 90s)
3. `message_part` (tool result) — client shows "Summarized" in stream
state
4. `message` (assistant) — durable message persisted, stream state
resets
5. `message` (tool) — durable tool result persisted
6. `status: waiting` — chat transitions to idle

## Tests

- **`OnStartFiresBeforePersist`**: Verifies callback ordering is
`on_start` → `generate` → `persist`.
- **`OnStartNotCalledBelowThreshold`**: Verifies `OnStart` is not called
when context usage is below the compaction threshold.

coderd/chatd/chatloop/compaction.go

@@ -30,6 +30,7 @@ type CompactionOptions struct {
 	SystemSummaryPrefix string
 	Timeout             time.Duration
 	Persist             func(context.Context, CompactionResult) error
+	OnStart             func()
 	OnError             func(error)
 }
 
@@ -134,6 +135,10 @@ func maybeCompact(
 		return nil
 	}
 
+	if config.OnStart != nil {
+		config.OnStart()
+	}
+
 	summary, err := generateCompactionSummary(
 		ctx,
 		runOpts.Model,

coderd/chatd/chatloop/compaction_test.go

@@ -83,6 +83,113 @@ func TestRun_Compaction(t *testing.T) {
 		require.InDelta(t, 80.0, persistedCompaction.UsagePercent, 0.0001)
 	})
 
+	t.Run("OnStartFiresBeforePersist", func(t *testing.T) {
+		t.Parallel()
+
+		const summaryText = "compaction summary for ordering test"
+
+		// Track the order of callbacks to verify OnStart fires
+		// before the Generate call (summary generation) and
+		// before Persist.
+		var callOrder []string
+
+		model := &loopTestModel{
+			provider: "fake",
+			streamFn: func(_ context.Context, _ fantasy.Call) (fantasy.StreamResponse, error) {
+				return streamFromParts([]fantasy.StreamPart{
+					{Type: fantasy.StreamPartTypeTextStart, ID: "text-1"},
+					{Type: fantasy.StreamPartTypeTextDelta, ID: "text-1", Delta: "done"},
+					{Type: fantasy.StreamPartTypeTextEnd, ID: "text-1"},
+					{
+						Type:         fantasy.StreamPartTypeFinish,
+						FinishReason: fantasy.FinishReasonStop,
+						Usage: fantasy.Usage{
+							InputTokens: 80,
+							TotalTokens: 85,
+						},
+					},
+				}), nil
+			},
+			generateFn: func(_ context.Context, _ fantasy.Call) (*fantasy.Response, error) {
+				callOrder = append(callOrder, "generate")
+				return &fantasy.Response{
+					Content: []fantasy.Content{
+						fantasy.TextContent{Text: summaryText},
+					},
+				}, nil
+			},
+		}
+
+		_, err := Run(context.Background(), RunOptions{
+			Model: model,
+			Messages: []fantasy.Message{
+				textMessage(fantasy.MessageRoleUser, "hello"),
+			},
+			MaxSteps: 1,
+			PersistStep: func(_ context.Context, _ PersistedStep) error {
+				return nil
+			},
+			ContextLimitFallback: 100,
+			Compaction: &CompactionOptions{
+				ThresholdPercent: 70,
+				SummaryPrompt:    "summarize now",
+				OnStart: func() {
+					callOrder = append(callOrder, "on_start")
+				},
+				Persist: func(_ context.Context, _ CompactionResult) error {
+					callOrder = append(callOrder, "persist")
+					return nil
+				},
+			},
+		})
+		require.NoError(t, err)
+		require.Equal(t, []string{"on_start", "generate", "persist"}, callOrder)
+	})
+
+	t.Run("OnStartNotCalledBelowThreshold", func(t *testing.T) {
+		t.Parallel()
+
+		onStartCalled := false
+
+		model := &loopTestModel{
+			provider: "fake",
+			streamFn: func(_ context.Context, _ fantasy.Call) (fantasy.StreamResponse, error) {
+				return streamFromParts([]fantasy.StreamPart{
+					{
+						Type:         fantasy.StreamPartTypeFinish,
+						FinishReason: fantasy.FinishReasonStop,
+						Usage: fantasy.Usage{
+							InputTokens: 10,
+						},
+					},
+				}), nil
+			},
+		}
+
+		_, err := Run(context.Background(), RunOptions{
+			Model: model,
+			Messages: []fantasy.Message{
+				textMessage(fantasy.MessageRoleUser, "hello"),
+			},
+			MaxSteps: 1,
+			PersistStep: func(_ context.Context, _ PersistedStep) error {
+				return nil
+			},
+			ContextLimitFallback: 100,
+			Compaction: &CompactionOptions{
+				ThresholdPercent: 70,
+				OnStart: func() {
+					onStartCalled = true
+				},
+				Persist: func(_ context.Context, _ CompactionResult) error {
+					return nil
+				},
+			},
+		})
+		require.NoError(t, err)
+		require.False(t, onStartCalled, "OnStart should not fire when usage is below threshold")
+	})
+
 	t.Run("ErrorsAreReported", func(t *testing.T) {
 		t.Parallel()