Add Flows usage table to GitLab Duo and SDLC trends dashboard

What does this MR do and why?

This change adds a new data table to GitLab's AI analytics dashboard that shows usage metrics for different AI agent workflow types. The table displays information like how many sessions each workflow had, average execution time, number of unique users, and completion rates. The data can be sorted by different columns, and the table is limited to showing the top 10 results. This gives teams visibility into how their AI agents are performing across different types of automated workflows.

References

Screenshots or screen recordings

Screen_Recording_2025-12-10_at_9.44.11_AM

How to set up and validate locally

• Apply this diff to mock the table data:

diff --git a/ee/app/graphql/resolvers/analytics/ai_metrics/agent_platform/flow_metrics_resolver.rb b/ee/app/graphql/resolvers/analytics/ai_metrics/agent_platform/flow_metrics_resolver.rb
index 4804d6679536..e9b82ca8f249 100644
--- a/ee/app/graphql/resolvers/analytics/ai_metrics/agent_platform/flow_metrics_resolver.rb
+++ b/ee/app/graphql/resolvers/analytics/ai_metrics/agent_platform/flow_metrics_resolver.rb
@@ -10,15 +10,18 @@ class FlowMetricsResolver < BaseResolver
           type [::Types::Analytics::AiMetrics::AgentPlatform::FlowMetricsType], null: true

           def resolve_with_lookahead
-            flow_metrics = ::Analytics::AiAnalytics::AgentPlatform::FlowMetricsService.new(
-              current_user,
-              namespace: context[:ai_metrics_namespace],
-              from: context[:ai_metrics_params][:start_date],
-              to: context[:ai_metrics_params][:end_date],
-              fields: lookahead.selections.map(&:name)
-            ).execute
+            # Mock data with 100 randomized records
+            flow_types = ['chat', 'code_review', 'fix_pipeline', 'explain_code', 'refactor_code']

-            flow_metrics.payload
+            100.times.map do
+              {
+                'flow_type' => flow_types.sample,
+                'sessions_count' => rand(1..5000),
+                'median_execution_time' => rand(5.0..300.0).round(1),
+                'users_count' => rand(1..500),
+                'completion_rate' => rand(30.0..100.0).round(1)
+              }
+            end
           end
         end
       end

• Navigate to the Duo and SDLC trends dashboard for any group/project
• Scroll down to the Flow usage table (below pipeline metrics)
• Data should render and be sortable by Number of sessions, Avg median duration and Unique users

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.

Related to #581554