Add retry logic with exponential backoff for IndexedDB write failures

[Copilot]

IndexedDB writes can fail transiently due to browser-level issues (transaction aborts, timeouts, blocking). Without retries, these failures result in silent data loss.

Changes

Retry Infrastructure

• Exponential backoff: 3 attempts max with ~1s, ~2s, ~4s delays (±10% jitter, capped at 5s)
• Error classification: Immediate failure for quota/permission errors; retry for transient failures (aborts, timeouts)
• Logging: Console output for retry attempts with timing and error details

Database Layer

Wrapped all write operations in database.js with retryOperation():

• Single/batch encrypted record storage
• User CRUD operations
• Metadata updates
• Migration operations

User Notifications

• Material-UI Snackbar context for non-retryable errors
• User-friendly messages: "Storage quota exceeded. Please free up space..." vs generic retry failures

Idempotency

Leverages Dexie's native put()/bulkPut() behavior—same ID updates existing record rather than creating duplicates.

Example

// Before: Direct IndexedDB write (no retry)
await db[storeName].put(record);

// After: Wrapped with retry logic
await retryOperation(
  async () => {
    await db[storeName].put(record);
  },
  `storeUserEncryptedRecord(${storeName}, ${id})`
);

Non-retryable quota errors fail immediately; transient errors retry with backoff. UI remains responsive via setTimeout() for async delays.

Testing

• 26 unit tests covering error classification, backoff calculation, retry logic
• 10 integration tests validating retry behavior with mocked Dexie operations
• Idempotency verified: multiple retries with same ID don't create duplicates

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Add retry logic for failed IndexedDB writes</issue_title>
<issue_description># Summary
Add best-effort retry logic for failed IndexedDB writes in v2 to reduce data loss from transient errors.

Requirements

• Wrap write ops with capped exponential backoff (e.g., ~1s, ~2s, ~4s; max 3 attempts).
• Classify errors: retry transient failures; surface non-retryable (e.g., quota exceeded) immediately with a clear UI message.
• Keep retries off the main thread; UI remains responsive.
• Ensure idempotency (no duplicate records on partial retries).
• Add lightweight logging/metrics for retry attempts and outcomes.

Acceptance Criteria

• Transient write failures are retried with capped backoff (up to 3 attempts).
• Non-retryable errors display a clear message (e.g., quota/storage full).
• No duplicate records created by the retry mechanism.
• UI stays responsive during retries; no freezes.
• Basic logs confirm retries and final status.</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Add retry logic for failed IndexedDB writes #120

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.