docs: document BatchTagPipeline ReadFn seam from PR #367

[HanSur94]

Summary

Wiki follow-up for #367 ("Tag-pipeline ReadFn seam, dashboard fixes, MISS_HIT lint cleanup").

That PR added a genuine new public feature — a 'ReadFn' constructor option on BatchTagPipeline, letting callers inject a custom raw-file parser (@(absPath)->parsed) for source formats the generic delimited reader can't handle. However, the PR never updated the class's own MATLAB docstring for this option, so scripts/generate_api_docs.py (which scrapes docstrings to build wiki/API-Reference:-Sensors.md) never picked it up — the auto-generated wiki page silently missed the new option and its new TagPipeline:invalidReadFn error.

(The rest of #367 — the FastSense/SparklineCardWidget/IconCardWidget empty-data crash fixes, the MISS_HIT lint cleanup, and the .gitignore chore — is internal robustness/tooling work with no public API surface, so no further wiki changes were needed there; the FastSense doc-comment changes in that PR were already picked up automatically in the post-merge docs: auto-update API reference from source code commit.)

Changes

• libs/SensorThreshold/BatchTagPipeline.m: document 'ReadFn' in the class header's error list and in the constructor's docstring (a third example line + a short explanation + the new error ID), matching the existing docstring conventions in this file.
• wiki/API-Reference:-Sensors.md: regenerated via python3 scripts/generate_api_docs.py (no other pages changed) — now shows the 'ReadFn' constructor example and the TagPipeline:invalidReadFn error for BatchTagPipeline.

Test plan

• Ran scripts/generate_api_docs.py and confirmed only wiki/API-Reference:-Sensors.md changed, with exactly the two intended lines added.
• Diffed all five generated API reference pages to confirm no incidental reformatting/drift.

🤖 Generated with Claude Code

https://claude.ai/code/session_01APr4Z5H8Ts437fhDtbyRYr

---

Generated by Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[HanSur94]

The MATLAB Tests (Q-Z) failure here is unrelated to this PR's diff (a MATLAB doc comment + a regenerated markdown file — no executable code changed). From the job log:

1. TestTagPerfRegression/testConsumerMigrationTickGate errored on a noisy timing gate: Tag-path tick overhead 25.1% exceeds 10% gate (legacy 567.5ms vs tag 709.9ms) — a flaky perf-regression benchmark, not a real regression.
2. MATLAB then crashed with a fatal error inside a MEX call while running TestZoomPan (segfault in libmwiqm.so/libmwbridge.so), which took down the whole batch and left .matlab-tests-passed unwritten.

Neither test touches BatchTagPipeline/SensorThreshold or the wiki. I tried to re-run just the failed jobs but my GitHub token doesn't have actions:write on this repo (403 Resource not accessible by integration). Could someone with write access re-run the failed jobs, or push an empty commit to re-trigger CI?

---

Generated by Claude Code