feat: replace Playwright extractor with yt-dlp subprocess

- Add instagram-extractor.ts: yt-dlp subprocess backend for Instagram
  caption extraction. No in-process browser state, maintained against
  Instagram frontend churn, supports cookies.txt for auth-walled reels.
- Add feature flag EXTRACTOR_BACKEND (ytdlp|playwright) in QueueProcessor
  so the old Playwright path remains available as fallback.
- Add 9 unit tests and 2 live-network integration tests for the new extractor.
- Dockerfile: install yt-dlp via pip3 alongside existing Chromium deps.
- docker-compose: expose EXTRACTOR_BACKEND env var (default: ytdlp).

Also in this commit:
- LLM: configurable per-request timeout via LLM_REQUEST_TIMEOUT_MS (default 120s);
  set maxRetries=0 to surface errors immediately; llama-swap /running health probe.
- QueueProcessor: thread progress callback through parser phase.
- LlmHealthIndicator: surface llama-swap loaded-model name.
- Logging: improve error serialization in queue-processor tests.
- .env.example: document llama-swap endpoint and model options.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

src/tests/queue-processor.spec.ts

@@ -35,13 +35,21 @@ vi.mock('$lib/server/queue/config', () => ({
 	}
 }));
 
-// Mock external dependencies BEFORE importing QueueProcessor
+// Mock external dependencies BEFORE importing QueueProcessor.
+// QueueProcessor.extractionPhase picks between two extractor modules based on
+// EXTRACTOR_BACKEND; mock both so behavior is identical regardless of default.
 vi.mock('$lib/server/extraction', () => ({
 	extractTextAndThumbnail: vi.fn().mockResolvedValue({
 		bodyText: 'Default recipe text',
 		thumbnail: null
 	})
 }));
+vi.mock('$lib/server/instagram-extractor', () => ({
+	extractTextAndThumbnail: vi.fn().mockResolvedValue({
+		bodyText: 'Default recipe text',
+		thumbnail: null
+	})
+}));
 
 vi.mock('$lib/server/parser', () => ({
 	extractRecipe: vi.fn().mockResolvedValue({
@@ -62,11 +70,16 @@ vi.mock('$lib/server/tandoor', () => ({
 	})
 }));
 
-import { extractTextAndThumbnail } from '$lib/server/extraction';
+import { extractTextAndThumbnail as extractFromExtraction } from '$lib/server/extraction';
+import { extractTextAndThumbnail as extractFromYtDlp } from '$lib/server/instagram-extractor';
 import { extractRecipe } from '$lib/server/parser';
 import { uploadRecipeWithIngredientsDTO, uploadRecipeImage } from '$lib/server/tandoor';
 import * as configModule from '$lib/server/queue/config';
 
+// Alias used by existing assertions; default backend is ytdlp so the new
+// instagram-extractor mock is what the processor actually invokes.
+const extractTextAndThumbnail = extractFromYtDlp;
+
 // Import processor AFTER mocks - it will auto-start (imported for side effects)
 import '$lib/server/queue/QueueProcessor';
 
@@ -78,8 +91,13 @@ describe('QueueProcessor Integration Tests', () => {
 		// Reset mocks and their implementations
 		vi.resetAllMocks();
 
-		// Set default mock implementations
-		vi.mocked(extractTextAndThumbnail).mockResolvedValue({
+		// Set default mock implementations on BOTH backend modules so the test
+		// behavior is invariant to EXTRACTOR_BACKEND.
+		vi.mocked(extractFromExtraction).mockResolvedValue({
+			bodyText: 'Default recipe text',
+			thumbnail: null
+		});
+		vi.mocked(extractFromYtDlp).mockResolvedValue({
 			bodyText: 'Default recipe text',
 			thumbnail: null
 		});