fix: resolve 500 errors from model returning trivial status objects (root cause + v0.2 prompt fix)

Two bugs were causing the model to return {"status":"ok"} / {"status":"ready"}
instead of structured reconstruction data, resulting in POST /api/analyse 500:

1. DOUBLE-WRAPPING BUG (lib/llm/provider.js):
   generateReconstruction() called buildPrompt(scenario) on input that was
   already a fully-built prompt string from analyseScenario(). This wrapped the
   v0.1 prompt (~5000+ chars) in another template layer, producing incomprehensible
   output that the model could not parse as structured JSON.
   Fix: Pass scenario through directly (it is ALREADY a built prompt).

2. MISSING JSON SPEC (prompts/reconstruct-v0.2.md):
   The v0.2 prompt template said 'matching the structure exactly' but never
   defined what that structure was. The model invented its own field names
   (input_classification, reasoning_mode, anchors) with snake_case instead of
   camelCase, which failed Zod validation -> 500 errors.
   Fix: Added explicit JSON schema section with exact key names, enum values,
   and nested structure matching the Zod validation layer.

Additionally:
- Refactored route to use analyseScenario from lib/analysis (centralized)
- Added lib/analysis.js with shared analysis logic
- Updated components to display promptVersion and validation errors
- Added lib/reconstruction/prompt.js v0.1/v0.2 versioning
- Added lib/reconstruction/schema.js v0.2 Zod schemas
- Added debug tool scripts, evaluation results, and comparison findings

lib/reconstruction/prompt.js

@@ -1,5 +1,17 @@
-export function buildPrompt(scenario) {
-  return `You are a neutral analyst performing an evidence-based reconstruction of the following scenario. 
+import { promises as fs } from "node:fs";
+import { fileURLToPath } from "node:url";
+import { dirname, join } from "node:path";
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = dirname(__filename);
+const PROMPTS_DIR = join(__dirname, "../../prompts");
+
+/** Available prompt versions */
+export const PROMPT_VERSIONS = ["v0.1", "v0.2"];
+
+/** Build a v0.1 (extraction-only) prompt inline for backward compatibility */
+function buildV1Prompt(scenario) {
+  return `You are a neutral analyst performing an evidence-based reconstruction of the following scenario.
 
 Rules:
 1. Do NOT invent facts. Only include information present in the scenario or clearly implied.
@@ -29,3 +41,34 @@ Return valid JSON matching this structure exactly:
 
 Return ONLY the JSON object. No markdown, no explanation, no preamble.`;
 }
+
+/** Load a versioned prompt from disk and substitute {{SCENARIO}} */
+async function buildV2Prompt(scenario) {
+  try {
+    const content = await fs.readFile(join(PROMPTS_DIR, "reconstruct-v0.2.md"), "utf-8");
+    return content.replace("{{SCENARIO}}", scenario);
+  } catch {
+    // Fall back to v0.1 prompt if v0.2 file is missing
+    return buildV1Prompt(scenario);
+  }
+}
+
+/**
+ * Build an analysis prompt for the given version.
+ * @param {"v0.1" | "v0.2"} [version="v0.2"]
+ * @returns {Promise<{prompt: string, version: string}>}
+ */
+export async function buildPrompt(scenario, version = "v0.2") {
+  let prompt;
+  switch (version) {
+    case "v0.1":
+      prompt = buildV1Prompt(scenario);
+      break;
+    default: // v0.2
+      prompt = await buildV2Prompt(scenario);
+      break;
+  }
+
+  const strongJsonHint = "\n\nReturn ONLY a valid JSON object starting with { and ending with }. Do NOT include any text before the opening brace or after the closing brace. Do NOT wrap in markdown backticks.";
+  return { prompt: prompt + strongJsonHint, version };
+}