feat: rebuild Experiments tab with Difference-in-Differences analysis and Bayesian verdicts

Replace the basic CRUD experiment tracker with a scientifically rigorous A/B testing system:

- Add DiD analysis engine (services/experimentAnalysis.ts) that computes treatment vs control
  group comparisons across 7 metrics (units, sessions, CVR, CTR, ROAS, revenue, ACOS)
- Implement Bayesian verdict system (Winner/Loser/Inconclusive) using posterior probability
  with normal CDF approximation (Abramowitz & Stegun erf, no external deps)
- Build counterfactual time series for trend charts (actual vs estimated without change)
- Rewrite ExperimentsView as single component with 3 inline sub-views (list, detail, create)
  replacing the previous modal-based ExperimentDetail and ExperimentForm
- Add control group support, change annotations (before→after diffs), and SEO experiment type
- New types: ExperimentChangeAnnotation, DiDMetricResult, DifferenceInDifferencesResult,
  ExperimentVerdict
- Simplify App.tsx by removing experiment modal state (5 useState hooks eliminated)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

types.ts

@@ -281,9 +281,31 @@ export interface VendorCSVRow {
 }
 
 // Experiment Tracking Types
-export type ExperimentType = 'pricing' | 'advertising' | 'content' | 'promotion';
+export type ExperimentType = 'pricing' | 'advertising' | 'content' | 'promotion' | 'seo';
 export type ExperimentStatus = 'planned' | 'active' | 'completed' | 'paused';
-export type ExperimentMetric = 'units' | 'revenue' | 'acos' | 'ctr' | 'cvr' | 'bsr';
+export type ExperimentMetric = 'units' | 'sessions' | 'cvr' | 'ctr' | 'roas' | 'revenue' | 'acos';
+export type ExperimentVerdict = 'winner' | 'loser' | 'inconclusive';
+
+export interface ExperimentChangeAnnotation {
+  field: string;
+  before_value: string;
+  after_value: string;
+}
+
+export interface DiDMetricResult {
+  treatment_before: number;
+  treatment_after: number;
+  control_before: number;
+  control_after: number;
+  did_estimate: number;
+  lift_percent: number;
+  posterior_prob_positive: number;
+}
+
+export interface DifferenceInDifferencesResult {
+  metrics: Record<string, DiDMetricResult>;
+  computed_at: string;
+}
 
 export interface Experiment {
   id: string;
@@ -292,6 +314,7 @@ export interface Experiment {
   type: ExperimentType;
   status: ExperimentStatus;
   asins: string[];
+  control_asins: string[];
   marketplace: string;
   start_date: string;
   end_date?: string;
@@ -300,23 +323,27 @@ export interface Experiment {
   hypothesis?: string;
   primary_metric: ExperimentMetric;
   target_lift_percent?: number;
+  changes: ExperimentChangeAnnotation[];
 
-  // Results
+  // Legacy results (kept for backward compat)
   baseline_units?: number;
   baseline_revenue?: number;
   baseline_gv?: number;
   baseline_cvr?: number;
   baseline_acos?: number;
-
   experiment_units?: number;
   experiment_revenue?: number;
   experiment_gv?: number;
   experiment_cvr?: number;
   experiment_acos?: number;
-
   actual_lift_percent?: number;
   statistical_significance?: number;
 
+  // DiD results
+  did_results?: DifferenceInDifferencesResult;
+  verdict?: ExperimentVerdict;
+  verdict_probability?: number;
+
   learnings?: string;
   owner?: string;
 }
@@ -326,12 +353,14 @@ export interface ExperimentCreateInput {
   description?: string;
   type: ExperimentType;
   asins: string[];
+  control_asins: string[];
   marketplace: string;
   start_date: string;
   end_date?: string;
   hypothesis?: string;
   primary_metric: ExperimentMetric;
   target_lift_percent?: number;
+  changes: ExperimentChangeAnnotation[];
   owner?: string;
 }
 
@@ -342,11 +371,14 @@ export interface ExperimentListItem {
   status: ExperimentStatus;
   asin_count: number;
   asins: string[];
+  control_asin_count: number;
   marketplace: string;
   start_date: string;
   end_date?: string;
   progress_percent: number;
   primary_metric: ExperimentMetric;
+  verdict?: ExperimentVerdict;
+  verdict_probability?: number;
   actual_lift_percent?: number;
   owner?: string;
 }