feat(issue summary) Only show possible cause if confident and novel (#84349)

## Background

`possible_cause_confidence` increases when there's less speculation
(which correlates w/ slightly more accurate causes)

`possible_cause_novelty` increases when there's more novelty / less
redundancy wrt `whats_wrong`

thresholds led to 50% of possible causes getting dropped on our autofix
sentry issues ([notebook
here](https://github.com/getsentry/data-analysis/blob/main/autofix/issue_summary/issue_summary_confidence.ipynb))

## Backend changes

corresponding backend change to sentry: #84346

corresponding backend change to seer:
getsentry/seer#1788

without the sentry backend change (there are no `data.scores`), default
to current behavior: always show possible cause


![before](https://github.com/user-attachments/assets/9f632a94-690b-4602-bdea-cbd2b95824b0)

with the change, only show it if both scores are greater than the
threshold


![after](https://github.com/user-attachments/assets/0b1e84af-cc0b-4d0a-8106-58cc615ea602)

(note: this example is actually considered novel and confident-enough. I
hardcoded the threshold to test the behavior)

Author: kddubey

static/app/components/group/groupSummary.spec.tsx

@@ -23,6 +23,22 @@ describe('GroupSummary', function () {
     trace: 'Test trace',
     possibleCause: 'Test possible cause',
     headline: 'Test headline',
+    scores: {
+      possibleCauseConfidence: 0.9,
+      possibleCauseNovelty: 0.8,
+    },
+  };
+
+  const mockSummaryDataWithLowScores = {
+    groupId: '1',
+    whatsWrong: 'Test whats wrong',
+    trace: 'Test trace',
+    possibleCause: 'Test possible cause',
+    headline: 'Test headline',
+    scores: {
+      possibleCauseConfidence: 0.5,
+      possibleCauseNovelty: 0.0,
+    },
   };
 
   beforeEach(() => {
@@ -62,6 +78,27 @@ describe('GroupSummary', function () {
     expect(screen.getByText('Test possible cause')).toBeInTheDocument();
   });
 
+  it('renders the summary without possible cause', async function () {
+    MockApiClient.addMockResponse({
+      url: `/organizations/${mockProject.organization.slug}/issues/${mockGroup.id}/summarize/`,
+      method: 'POST',
+      body: mockSummaryDataWithLowScores,
+    });
+
+    render(<GroupSummary event={mockEvent} group={mockGroup} project={mockProject} />, {
+      organization,
+    });
+
+    await waitFor(() => {
+      expect(screen.getByText("What's wrong")).toBeInTheDocument();
+    });
+    expect(screen.getByText('Test whats wrong')).toBeInTheDocument();
+    expect(screen.getByText('In the trace')).toBeInTheDocument();
+    expect(screen.getByText('Test trace')).toBeInTheDocument();
+    expect(screen.queryByText('Possible cause')).not.toBeInTheDocument();
+    expect(screen.queryByText('Test possible cause')).not.toBeInTheDocument();
+  });
+
   it('shows loading state', function () {
     MockApiClient.addMockResponse({
       url: `/organizations/${mockProject.organization.slug}/issues/${mockGroup.id}/summarize/`,
@@ -73,8 +110,8 @@ describe('GroupSummary', function () {
       organization,
     });
 
-    // Should show loading placeholders
-    expect(screen.getAllByTestId('loading-placeholder')).toHaveLength(2);
+    // Should show loading placeholders. Currently we load the whatsWrong section
+    expect(screen.getAllByTestId('loading-placeholder')).toHaveLength(1);
   });
 
   it('shows error state', async function () {

static/app/components/group/groupSummary.tsx

@@ -16,11 +16,18 @@ import {useFeedbackForm} from 'sentry/utils/useFeedbackForm';
 import useOrganization from 'sentry/utils/useOrganization';
 import {useAiConfig} from 'sentry/views/issueDetails/streamline/hooks/useAiConfig';
 
+const POSSIBLE_CAUSE_CONFIDENCE_THRESHOLD = 0.468;
+const POSSIBLE_CAUSE_NOVELTY_THRESHOLD = 0.419;
+
 interface GroupSummaryData {
   groupId: string;
   headline: string;
   eventId?: string | null;
   possibleCause?: string | null;
+  scores?: {
+    possibleCauseConfidence: number;
+    possibleCauseNovelty: number;
+  } | null;
   trace?: string | null;
   whatsWrong?: string | null;
 }
@@ -152,6 +159,11 @@ export function GroupSummary({
       : []),
   ];
 
+  const shouldShowPossibleCause =
+    !data?.scores ||
+    (data.scores.possibleCauseConfidence >= POSSIBLE_CAUSE_CONFIDENCE_THRESHOLD &&
+      data.scores.possibleCauseNovelty >= POSSIBLE_CAUSE_NOVELTY_THRESHOLD);
+
   const insightCards = [
     {
       id: 'whats_wrong',
@@ -167,13 +179,17 @@ export function GroupSummary({
       icon: <IconSpan size="sm" />,
       showWhenLoading: false,
     },
-    {
-      id: 'possible_cause',
-      title: t('Possible cause'),
-      insight: data?.possibleCause,
-      icon: <IconFocus size="sm" />,
-      showWhenLoading: true,
-    },
+    ...(shouldShowPossibleCause
+      ? [
+          {
+            id: 'possible_cause',
+            title: t('Possible cause'),
+            insight: data?.possibleCause,
+            icon: <IconFocus size="sm" />,
+            showWhenLoading: false,
+          },
+        ]
+      : []),
   ];
 
   return (

static/app/views/issueDetails/streamline/sidebar/solutionsSection.spec.tsx

@@ -116,7 +116,7 @@ describe('SolutionsSection', () => {
     );
 
     expect(screen.getByText('Solutions Hub')).toBeInTheDocument();
-    expect(screen.getAllByTestId('loading-placeholder')).toHaveLength(3);
+    expect(screen.getAllByTestId('loading-placeholder')).toHaveLength(2); // whatsWrong and Open Autofix
   });
 
   it('renders summary when AI features are enabled and data is available', async () => {