Parallelize GitHub data cache generation

[neubig]

This PR improves the performance of the GitHub data cache generation by parallelizing the comment fetching process.

Changes:

• Process items in parallel batches of 10 to avoid rate limiting
• Use Promise.all to fetch comments for multiple issues/PRs simultaneously
• Add better progress reporting showing which batch is being processed

Performance improvements:

• Initial fetch: 4.70s (was 4.82s)
• Processing items: 12.40s (was 86.83s, 7x faster!)
• Sorting: 0.00s (unchanged)
• Total time: 17.11s (was 91.65s, 5.4x faster overall!)

[vercel]

@openhands-agent is attempting to deploy a commit to the all-hands-ai team on Vercel, but is not a member of this team. To resolve this issue, you can:

• Make your repository public. Collaboration is free for open source and public repositories.
• Add @openhands-agent as a member. A Pro subscription is required to access Vercel's collaborative features.
  • If you're the owner of the team, click here and add @openhands-agent as a member.
  • If you're the user who initiated this build request, click here to request access.
  • If you're already a member of the all-hands-ai team, make sure that your Vercel account is connected to your GitHub account.

To read more about collaboration on Vercel, click here.