usage: continuous Ollama Cloud usage sampling + delta report

- scripts/ollama_usage_poll.py: per-minute cron sample into db/ollama_usage.sqlite,
  write-on-change; meta table records every poll so a data gap can be told apart
  from a failed or missed poll
- scripts/ollama_usage_report.py: delta report keyed on per-model request_count
  (limits.*.usage has 0.1 % resolution, short-interval deltas are noise)
- SKILL.md: Continuous sampling section

skills/usage/SKILL.md

@@ -46,6 +46,22 @@ UTC. Rationale: session usage climbed in real time during testing (rolling
 window), weekly changes slowly — consistent with hourly/weekly windows.
 If the window turns out not to be calendar-based, fix `until_next_*`.
 
+## Continuous sampling
+
+A cron job runs `scripts/ollama_usage_poll.py` every minute and appends to
+`db/ollama_usage.sqlite` whenever anything changed (table `samples`; table `meta`
+records every poll, so a gap can be told apart from a failed poll).
+
+For a delta report over that data:
+
+```bash
+uv run skills/usage/scripts/ollama_usage_report.py [--since ISO] [--until ISO]
+```
+
+Default window is the last 24 hours. Per-model **request counts** are the exact
+figure there — `limits.*.usage` has a resolution of 0.1 %, so short-interval
+percentage deltas are noise.
+
 ## Notes
 
 - Endpoint: `GET https://ollama.com/api/usage`, header