Add failed-upload retention cleanup and legal warning banners

[wickedyoda]

Summary

• keep partial files for failed uploads for 2 minutes before cleanup
• add POST /api/upload/fail/:uploadId to mark failed uploads for delayed cleanup
• add periodic failed-upload cleanup worker and env toggle (DISABLE_FAILED_UPLOAD_CLEANUP)
• update upload client to mark failed uploads instead of immediate cancellation on non-user errors
• add top and bottom legal warning/disclaimer links on upload page
• add regression test for failed upload retention window and cleanup timing

Verification

• static error check in edited files: no errors
• runtime tests could not be executed in this environment because node is not available (command not found)

[Copilot]

Pull request overview

Adds a failed-upload retention flow (mark-as-failed endpoint + periodic cleanup) and surfaces legal warning banners on the upload page, with a regression test for retention timing.

Changes:

• Introduces POST /api/upload/fail/:uploadId, a failed-upload retention window constant, and a periodic cleanup worker gated by DISABLE_FAILED_UPLOAD_CLEANUP.
• Updates the upload client to mark uploads as failed on non-user errors (instead of immediate server cancellation).
• Adds top/bottom legal warning banners to the upload UI, plus a test validating retention and cleanup behavior.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
src/routes/upload.js	Adds failed-upload retention state, cleanup worker, and /fail endpoint; clears failed marker on resumed chunk uploads.
test/upload.test.js	Adds regression test validating failed-upload retention and cleanup timing via exported helpers.
public/index.html	Adds legal warning banners and switches client failure handling to call the new /fail endpoint.
public/styles.css	Styles the new legal warning banners and adjusts responsive sizing.
.env.example	Documents DISABLE_FAILED_UPLOAD_CLEANUP toggle.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

In the chunk handler, metadata.failedAt is cleared in-memory but isn’t persisted until after the chunk write. The failed-upload cleanup interval can race with an upload resuming and delete the partial file/metadata during this window (worst case: the file is deleted and then re-created by fs.open(..., 'a'), corrupting the upload). Persist the cleared failedAt (and/or set an explicit active state) before any file operations, and/or have the cleanup worker also gate deletion on lastActivity being older than the retention window.

Suggested change

	// Upload resumed after failure; clear failed marker so cleanup won't remove active data.
	delete metadata.failedAt;
	// Upload resumed after failure; clear failed marker and persist activity
	// before any file operations so cleanup logic cannot race and delete
	// the partial upload based on stale on-disk metadata.
	delete metadata.failedAt;
	metadata.lastActivity = new Date().toISOString();
	await writeUploadMetadata(uploadId, metadata);

[Copilot]

The upload routes module now starts the failed-upload cleanup interval on import. This test file disables batch cleanup, but doesn’t disable failed-upload cleanup, so the background interval can make the new retention test flaky on slow runs (it may delete the partial/metadata based on wall clock time). Consider setting process.env.DISABLE_FAILED_UPLOAD_CLEANUP = 'true' in tests (and/or calling stopFailedUploadCleanup() in teardown) and invoking cleanupFailedUploads(...) explicitly when needed.