Improve active_learning density and checkpoint quality in 08b-interpret-your-run.md

[Copilot]

08b-interpret-your-run.md scored below benchmark on active_learning (density 0.88 vs ≥3.0/100 words) and checkpoint_quality (3 items vs ≥4 minimum), making it a passive reading exercise with no structured verification.

Changes

• Structured fill-in exercise — replaces the vague "write one short note" prompt with a labeled code-fence template including example values, giving learners a concrete scaffold to verify they read the log correctly:

  First Tool call I saw:         [tool name, e.g. github.list_issues]
  What it was trying to do:      [one sentence description]
  

• Failed run log example — adds a second text code block in the error-patterns section showing a 403 Forbidden failure signature, with a context sentence explaining it helps distinguish permission errors from vague-brief errors at a glance

• Checkpoint expanded to 4 items — adds: "I can identify whether a run failed due to a permission error, a vague brief, or a missing output", directly anchoring the three error patterns listed above it

File stays at 571 words (limit: 600).

• Fixes [curriculum-eval] 08b-interpret-your-run.md: active_learning — activity density 0.88, only 1 code block and 3 checklist items #1776