Break out GitHub Packages downloads per tag

[MikeMcQuaid]

brew info --github-packages-downloads printed a single 30 day total. Show a table per GitHub Packages tag with a grand total instead, in the same table_output format as the other analytics tables, and add the signed percentage difference from the 30 day analytics install events so under-reporting is visible at a glance:

==> github-packages-downloads (30 days)
Index | Tag      |  Count |  Percent
-----:|----------|-------:|--------:
1     | 1.25.0-2 | 99,129 |   99.26%
2     | 1.25.0   |    427 |    0.43%
3     | 1.25.0-1 |    263 |    0.26%
4     | 1.24.5   |     32 |    0.03%
5     | 1.25.0_1 |     15 |    0.02%
Total |          | 99,866 |  100.00%
Difference from 16,181 analytics install events (30 days): +517.18%

A per-platform (bottle tag, OS or architecture) breakdown is not possible from GitHub Packages. GitHub counts downloads per version, i.e. per manifest. Homebrew fetches the multi-arch index by version tag (manifests/1.25.0-2) and then the bottle blob by digest: blobs are not versions so they are never counted, and the per-platform child manifests only see a handful of lifetime downloads from tools like docker or oras (10-135 each versus ~100k per month on the tagged index above). Per-platform install data only exists in the Influx analytics os and arch tags.

• Fetch all tag pages with one curl --parallel --compressed call into a temporary directory instead of one uncompressed curl process per tag. Five pages take ~0.12s instead of ~2.2s (33KB versus 210KB each, over a single HTTP/2 connection); the regex parsing was never the cost.

---

• Have you followed our Contributing guidelines?
• Have you checked for other open Pull Requests for the same change?
• Have you explained what your changes do? Performance claims (e.g. "this is faster") must include brew benchmark results.
• Have you explained why you'd like these changes included, not just what they do?
• For bug fixes, have you given step-by-step brew commands to reproduce the bug?
• Have you written new tests (excluding integration tests)? Here's an example.
• Have you successfully run brew lgtm (style, typechecking and tests) locally?

---

• I did not use AI/LLM to create this PR, or I disclosed the tool/model below and reviewed its output; I did not attribute commits to AI and will answer maintainer questions and review comments myself without AI/LLM.

Claude with Fable at Extra High effort, with local review and testing.

---

[Copilot]

Pull request overview

Adds per-tag GitHub Packages download reporting and compares totals with 30-day install analytics.

Changes:

• Downloads tag pages concurrently with compression.
• Displays sorted per-tag counts and analytics percentage difference.
• Adds coverage for parsing and table output.

Reviewed changes

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

File	Description
Library/Homebrew/utils/analytics.rb	Implements parallel per-tag reporting and custom table headers.
Library/Homebrew/test/utils/analytics_spec.rb	Tests per-tag counts, totals, and analytics comparison.

---

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.