fix: batch optimization stage 1 (#80)

* Feat: Implement batch processing and optimized article insertion for RSS fetching

* perf: increase feed batch size and add queue implementation plan

- Increase maxFeedsPerBatch from 20 to 100 for faster rotation
- With 1-minute cron frequency, this processes 6,000 feeds/hour
- Add comprehensive queue-based architecture plan for Phase 3
- Plan includes runtime abstraction for both Cloudflare and Docker
- Supports scaling from 2K to 50K+ feeds with <5min freshness

Queue plan features:
- Cloudflare Queues for Workers deployment
- BullMQ (Redis) for Docker deployment
- In-memory fallback for development
- Graceful degradation when queue unavailable
- Cost projections and migration strategy included

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* chore: format

* fix: address Sentry bug and ESLint errors

Critical fixes:
- Use .limit() server-side instead of loading all feeds into memory
  This prevents memory pressure (e.g., 25MB for 50K feeds) and scales better
- Add separate lightweight query for total count (only fetches IDs)
- Fix ESLint error: add proper type assertion for D1 batch statements

Technical details:
- Changed from allSources.slice() to db.select().limit()
- Added comment explaining Drizzle/D1 type mismatch
- Formatted queue implementation plan with Prettier

Note: Existing type errors with inArray() remain unfixed (pre-existing issue)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* fix: resolve type errors from batch optimization commit

The batch optimization commit (913b220) introduced type errors by using
.select() with field specifications, which breaks Drizzle ORM's type
inference for inArray().

Fixes:
- Line 205: Changed .select({ plan }) back to .select()
- Line 560: Changed .select({ guid }) back to .select()
- Line 92: Changed .select({ id }) back to .select()

Root cause: Drizzle's inArray() requires full table selection for proper
type inference. Using partial field selection breaks the type chain.

Confirmed: main branch has no type errors, errors were introduced in 913b220

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* chore: format

* chore: simplify format scripts in monorepo

Replace complex multi-package format scripts with single prettier
command that formats entire repository in one pass.

Before:
- format:root for root files
- Separate format commands for each package
- 4 total script definitions

After:
- Single `prettier --write .` command
- Leverages .prettierignore for exclusions
- Simpler, faster, and more maintainable

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* chore: format all files with prettier

Run prettier across entire codebase to fix trailing newlines,
line wrapping, and formatting inconsistencies.

Changes include:
- Remove trailing newlines from config files
- Fix line wrapping in markdown documentation
- Standardize formatting in YAML workflow files
- Reformat JSON snapshot files
- Apply consistent code formatting across packages

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* chore: format

---------

Co-authored-by: seer-by-sentry[bot] <157164994+seer-by-sentry[bot]@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>

Author: 3 people

.github/ISSUE_TEMPLATE/config.yml

@@ -3,4 +3,3 @@ contact_links:
   - name: Documentation
     url: https://github.com/TechSquidTV/Tuvix-RSS/tree/main/docs
     about: Check out the comprehensive documentation
-

.github/actions/get-worker-name/action.yml

@@ -1,23 +1,23 @@
-name: 'Get Worker Name'
-description: 'Extracts worker name from wrangler.toml and optionally appends suffix'
+name: "Get Worker Name"
+description: "Extracts worker name from wrangler.toml and optionally appends suffix"
 
 inputs:
   wrangler-toml-path:
-    description: 'Path to wrangler.toml file'
+    description: "Path to wrangler.toml file"
     required: false
-    default: 'packages/api/wrangler.toml'
+    default: "packages/api/wrangler.toml"
   suffix:
-    description: 'Optional suffix to append to worker name (e.g., -dev)'
+    description: "Optional suffix to append to worker name (e.g., -dev)"
     required: false
-    default: ''
+    default: ""
 
 outputs:
   name:
-    description: 'The worker name'
+    description: "The worker name"
     value: ${{ steps.get-name.outputs.name }}
 
 runs:
-  using: 'composite'
+  using: "composite"
   steps:
     - name: Get worker name
       id: get-name
@@ -28,4 +28,3 @@ runs:
         echo "name=$WORKER_NAME" >> $GITHUB_OUTPUT
         echo "WORKER_NAME=$WORKER_NAME" >> $GITHUB_ENV
         echo "Worker name: $WORKER_NAME"
-

.github/actions/setup/action.yml

@@ -1,18 +1,18 @@
-name: 'Setup Node.js and pnpm'
-description: 'Sets up Node.js and pnpm with caching'
+name: "Setup Node.js and pnpm"
+description: "Sets up Node.js and pnpm with caching"
 
 inputs:
   node-version:
-    description: 'Node.js version'
+    description: "Node.js version"
     required: false
-    default: '20'
+    default: "20"
   pnpm-version:
-    description: 'pnpm version'
+    description: "pnpm version"
     required: false
-    default: '10.19.0'
+    default: "10.19.0"
 
 runs:
-  using: 'composite'
+  using: "composite"
   steps:
     - name: Setup pnpm
       uses: pnpm/action-setup@v4
@@ -23,9 +23,8 @@ runs:
       uses: actions/setup-node@v4
       with:
         node-version: ${{ inputs.node-version }}
-        cache: 'pnpm'
+        cache: "pnpm"
 
     - name: Install dependencies
       shell: bash
       run: pnpm install --frozen-lockfile
-

.github/actions/substitute-d1-database-id/action.yml

@@ -1,21 +1,21 @@
-name: 'Create wrangler.toml from example'
-description: 'Creates wrangler.toml from wrangler.example.toml and substitutes D1_DATABASE_ID'
+name: "Create wrangler.toml from example"
+description: "Creates wrangler.toml from wrangler.example.toml and substitutes D1_DATABASE_ID"
 
 inputs:
   wrangler-example-path:
-    description: 'Path to wrangler.example.toml file'
+    description: "Path to wrangler.example.toml file"
     required: false
-    default: 'packages/api/wrangler.example.toml'
+    default: "packages/api/wrangler.example.toml"
   wrangler-toml-path:
-    description: 'Path where wrangler.toml will be created'
+    description: "Path where wrangler.toml will be created"
     required: false
-    default: 'packages/api/wrangler.toml'
+    default: "packages/api/wrangler.toml"
   d1-database-id:
-    description: 'D1 Database ID secret'
+    description: "D1 Database ID secret"
     required: true
 
 runs:
-  using: 'composite'
+  using: "composite"
   steps:
     - name: Create wrangler.toml from example
       shell: bash

.github/workflows/CI_IMPROVEMENTS.md

@@ -9,12 +9,14 @@ This document outlines the improvements made to the CI/CD workflows to make them
 ### 1. DRY (Don't Repeat Yourself) ✅
 
 **Before:**
+
 - Setup steps (checkout, pnpm setup, node setup, install) repeated in every job
 - Worker name extraction logic duplicated across deployment workflows
 - D1 database ID substitution logic duplicated
 - Deployment version extraction duplicated
 
 **After:**
+
 - Created reusable composite actions:
   - `.github/actions/setup/` - Centralized Node.js and pnpm setup
   - `.github/actions/substitute-d1-database-id/` - Reusable D1 substitution logic
@@ -25,42 +27,51 @@ This document outlines the improvements made to the CI/CD workflows to make them
 ### 2. Efficiency Improvements ✅
 
 **Parallelization:**
+
 - All CI jobs (lint, format, type-check, test-api, test-app, build) already run in parallel
 - Tests run independently and can complete faster than sequential execution
 
 **Caching:**
+
 - pnpm cache is automatically handled by `actions/setup-node@v4` with `cache: 'pnpm'`
 - This significantly speeds up dependency installation on subsequent runs
 
 **Reduced Redundancy:**
+
 - CI workflows now use a single reusable workflow instead of duplicating steps
 - Deployment workflows share common actions
 
 ### 3. Quality Improvements ✅
 
 **Fixed Format Check:**
+
 - **Before:** `format:app` didn't use `--check` flag, so it would modify files instead of failing
 - **After:** Added `--check` flag to `format:app` script so CI properly fails on formatting issues
 
 **Removed Test Failure Masking:**
+
 - **Before:** Tests used `|| true` which masked failures
 - **After:** Removed `|| true` so test failures properly fail the CI
 
 **Proper Error Handling:**
+
 - Format checks now properly fail when code is not formatted
 - Test failures are no longer hidden
 
 ### 4. Speed Improvements ✅
 
 **Faster Setup:**
+
 - Reusable actions reduce workflow parsing time
 - Cached dependencies speed up installation
 
 **Parallel Execution:**
+
 - All jobs run in parallel (no dependencies between lint, format, type-check, tests)
 - Build job runs independently
 
 **Reduced Workflow Size:**
+
 - Smaller workflow files are faster to parse and validate
 - Less YAML to process
 
@@ -95,7 +106,7 @@ jobs:
     uses: ./.github/workflows/ci-reusable.yml
     with:
       upload_coverage: true
-      upload_artifacts: true  # Only for development branch
+      upload_artifacts: true # Only for development branch
 ```
 
 ### Deployment Workflows
@@ -114,7 +125,7 @@ Deployment workflows now use composite actions:
 - name: Get worker name
   uses: ./.github/actions/get-worker-name
   with:
-    suffix: '-dev'  # Optional
+    suffix: "-dev" # Optional
 ```
 
 ## Benefits
@@ -148,7 +159,7 @@ To verify the improvements work:
 4. Create a release → Should trigger `deploy-cloudflare.yml`
 
 All workflows should:
+
 - Complete faster due to parallelization
 - Fail properly on formatting/test issues
 - Use the new reusable components
-