Cal drift: track cable-loss .s2p history alongside cal runs

[RFingAdam]

Background

Cable insertion loss is a common source of chamber drift (connector wear, cable flex fatigue). The active-cal drift tracker records the generated TRP Cal file + reference-file metadata, but it does not currently capture cable-loss measurements (.s2p Touchstone files) that operators run during a full cal session.

Goal

Extend drift tracking so .s2p cable-loss files measured during or alongside an active cal are ingested as a third cal type, with their own compare + plot workflow.

Acceptance Criteria

• Parser for .s2p Touchstone format added to plot_antenna/cal_drift.py (S21 magnitude over frequency; dB or linear input both accepted).
• .s2p files appear as cal_type = "cable_loss" rows in runs.csv.
• GUI dialog can ingest a directory of .s2p files and compare any two for drift (ΔdB vs frequency, single-pol plot — no H/V split).
• MCP cal_drift_ingest recognizes .s2p files alongside TRP Cal *.txt.
• Tests: fixture .s2p file + parse/compare tests.
• At least one real-world .s2p pair (when available) ingests cleanly and produces a drift plot.

Notes

• User noted they do not currently have a consistent archive of historical .s2p files — this may need to start building an archive from the next cal onward. Prioritize accordingly.
• Consider: should cable-loss drift automatically subtract from active-cal drift numbers? That is a debatable correction — capture the raw numbers first, leave correction as a separate flag/feature.

Related: v1 tracker at plot_antenna/cal_drift.py. Existing RF tool scikit-rf is not currently a dependency; evaluate whether to add it vs. hand-roll a minimal Touchstone parser.