feat(prompt-optimization): implement Data Layer MVP and Tool Alignment dataset

Established the "Heart" of the Prompt Optimization Pipeline by building a robust,
extensible data infrastructure and a high-fidelity golden dataset.

Key improvements:
- Core Schema: Defined the `Scenario` interface in `data/schema.ts` supporting
  multiple negative failure modes, platform-specific shell contexts (Unix/Win32),
  and strict tool-call typing.
- Optimization Manifest: Created `data/manifest.json` to define "No-Fly Zones"
  for the optimizer, protecting literal tool names and template variables, while
  providing descriptive context for validation.
- Tool Alignment Dataset: Authored 113 scenarios in `data/tool_alignment.jsonl`
  across 20 tools, focusing on "Built-in over Shell" preference. Heavily weighted
  `replace` (12) and `write_file` (10) to enforce surgical editing.
- Extensible Validator: Implemented `scripts/validate-data.ts` to provide
  real-time integrity checks and purpose-driven coverage reports.
- Project Integration: Added `data:validate`, `data:format`, and `data:lint`
  scripts to package.json and updated ESLint config to cover the data directory.

data/schema.ts

@@ -0,0 +1,49 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+/**
+ * The core data interface for the Tool Alignment Dataset.
+ * Designed to be extensible for custom error reports and metrics.
+ */
+
+export interface ToolCall {
+  name: string;
+  arguments: Record<string, unknown>;
+}
+
+export interface NegativeExample {
+  id?: string;
+  tool_calls: ToolCall[];
+  output_text?: string; // For "too chatty" or "hallucination" failures
+  reason: string; // e.g., "Defaulted to shell 'cat'", "Included conversational filler"
+  severity: 'low' | 'medium' | 'high'; // Helps the optimizer prioritize fixes
+}
+
+export interface Scenario {
+  id: string; // Unique identifier (e.g., 'read_file-01')
+  metadata: {
+    tags: string[]; // e.g., ['tool-alignment', 'shell-avoidance']
+    created_at: string;
+    platform?: 'darwin' | 'linux' | 'win32'; // To handle platform-specific shell variations
+    model_info?: {
+      // Placeholder for future tracking
+      name?: string;
+      version?: string;
+    };
+  };
+  input: {
+    user_query: string;
+    context?: {
+      current_file?: string;
+      directory_structure?: string[];
+    };
+  };
+  expected: {
+    tool_calls: ToolCall[];
+    rationale: string; // Why this is the 'Golden' choice
+  };
+  negatives: NegativeExample[]; // Array of multiple failure modes
+}