Fix language runtime size measurement correctness (Issue #55)

[konard]

Summary

Fixes #55 — checks correctness of language runtime size measurements and fixes multiple root causes of incorrect/zero values.

Root Causes Found & Fixed

1. Integer Division Truncation → Replaced with Float Precision (Primary Fix)

Before: diff_bytes / 1024 / 1024 — floor division truncated any component under 1 MiB to 0 MB
After: awk "BEGIN {printf \"%.2f\", $diff_bytes / 1000000}" — precise decimal values

Affected components: Essential Tools (741 KB shows as 0.74 MB), GitHub CLI (549 KB shows as 0.55 MB), and improved accuracy across all components (e.g. 504905728 bytes shows as 504.91 MB instead of 505).

2. Unreliable df Delta for Rust and Homebrew

Both rustup and Homebrew use mv/hardlink operations internally that don't change filesystem block counts during installation, causing near-zero df deltas (~96 KB for Rust, ~16 KB for Homebrew) despite actual sizes of 700+ MB and 200+ MB respectively.

Fix: Switched to du -sb measurement of the actual installed directories after installation (instead of df delta).

3. Missing Component

bubblewrap is installed in ubuntu/24.04/full-sandbox/install.sh but was not measured. Added measurement.

4. Data & README Updated

• data/disk-space-measurements.json updated with corrected size_mb float values (e.g. 0.74 instead of 1)
• README.md component sizes table updated with precise float values
• Rust and Homebrew will show corrected sizes after next CI run (current JSON reflects old measurement data for those two)

Case Study

Full root cause analysis documented in docs/case-studies/issue-55/CASE-STUDY.md with timeline, evidence, and proposed solutions.

Changes

• scripts/measure-disk-space.sh — Use awk-based float division for size_mb, update add_measurement to store float, add bubblewrap, use du-based measurement for Rust and Homebrew
• data/disk-space-measurements.json — Corrected size_mb float values
• README.md — Updated component sizes table with precise decimal values
• docs/case-studies/issue-55/CASE-STUDY.md — New case study

🤖 Generated with Claude Code

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $2.855204
• Calculated by Anthropic: $2.510274 USD
• Difference: $-0.344930 (-12.08%)
  📎 Log file uploaded as Gist (2682KB)
  🔗 View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag

[konard]

(diff_bytes + 500000) I think that is wrong, we need to have precise calculations, so we will get something like 0.72 MB for 741 KB (note that this is just an example, and calculation may be incorrect).

[konard]

🤖 AI Work Session Started

Starting automated work session at 2026-02-23T08:47:21.687Z

The PR has been converted to draft mode while work is in progress.

This comment marks the beginning of an AI work session. Please wait for the session to finish, and provide your feedback.

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $2.135613
• Calculated by Anthropic: $1.850424 USD
• Difference: $-0.285189 (-13.35%)
  📎 Log file uploaded as Gist (2044KB)
  🔗 View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

🔄 Auto-restart 1/3

Detected uncommitted changes from previous run. Starting new session to review and commit them.

Uncommitted files:

?? SUMMARY.md
?? issue_comments.txt
?? issue_view.txt
?? pr_conversation_comments.txt
?? pr_diff.txt
?? pr_inline_comments.txt
?? pr_reviews.txt
?? pr_view.txt

---

Auto-restart will stop after changes are committed or after 2 more iterations. Please wait until working session will end and give your feedback.

[konard]

🔄 Auto-restart 1/3 Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $0.461826
• Calculated by Anthropic: $0.280741 USD
• Difference: $-0.181085 (-39.21%)
  📎 Log file uploaded as Gist (2659KB)
  🔗 View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.