P1: file parsers, sample sheet, discovery, CLI inspect

Readers for IDAT v3, BPM (locus entries v4-v8), CSV manifest, EGT
(data blocks 5-9) and GTC v3-v5 (plus a layout-preserving GTC writer);
sample-sheet validation (IN-001..IN-014), IDAT discovery and pairing
(IN-020..IN-022), message catalogue and exit codes (CLI-014/016),
JSON Lines logging (CLI-007/008), atomic writes (CLI-011), run manifest,
and `simurg inspect`.

Validation (docs/validation/P1.md): GTC, BPM and EGT readers agree
exactly with Illumina BeadArrayFiles on GSA v4 / GDA v1 files; GTCs
round-trip byte-identically; BPM and CSV manifests are consistent.

tests/unit/test_cli_errors.py

@@ -0,0 +1,95 @@
+"""CLI behaviour (CLI-007, CLI-014, CLI-020) and the message catalogue (CLI-016)."""
+
+from __future__ import annotations
+
+import json
+import subprocess
+import sys
+from pathlib import Path
+
+import numpy as np
+from typer.testing import CliRunner
+
+from simurg.cli.main import app
+from simurg.core.errors import CATALOGUE, render_markdown
+from simurg.synth.files import SynthIdat, SynthLocus, encode_bpm, encode_egt, encode_idat
+
+ROOT = Path(__file__).resolve().parents[2]
+runner = CliRunner()
+
+
+def _files(tmp_path: Path) -> dict[str, Path]:
+    idat = tmp_path / "200000000001_R01C01_Grn.idat"
+    idat.write_bytes(
+        encode_idat(SynthIdat(address_ids=np.arange(1, 6, dtype=np.int32), mean=np.arange(5, dtype=np.uint16)))
+    )
+    bpm = tmp_path / "m.bpm"
+    bpm.write_bytes(encode_bpm([SynthLocus("rs1"), SynthLocus("rs2", address_a=2)]))
+    egt = tmp_path / "c.egt"
+    egt.write_bytes(encode_egt(["rs1"], theta_mean=[[0.1, 0.5, 0.9]], r_mean=[[1, 1, 1]]))
+    return {"idat": idat, "bpm": bpm, "egt": egt}
+
+
+def test_version() -> None:
+    r = runner.invoke(app, ["--version"])
+    assert r.exit_code == 0 and r.stdout.startswith("SimurgArray ")
+
+
+def test_inspect_json(tmp_path) -> None:  # type: ignore[no-untyped-def]
+    f = _files(tmp_path)
+    r = runner.invoke(app, ["inspect", "--json", str(f["idat"]), str(f["bpm"]), str(f["egt"])])
+    assert r.exit_code == 0, r.output
+    out = json.loads(r.stdout)
+    assert [o["kind"] for o in out] == ["idat", "bpm", "egt"]
+    assert out[1]["num_loci"] == 2 and out[2]["num_records"] == 1
+
+
+def test_inspect_missing_file_exit2() -> None:
+    r = runner.invoke(app, ["inspect", "/nonexistent/x.idat"])
+    assert r.exit_code == 2
+    assert "E_FILE_NOT_FOUND" in r.output
+
+
+def test_inspect_partial_failure_exit3(tmp_path) -> None:  # type: ignore[no-untyped-def]
+    f = _files(tmp_path)
+    bad = tmp_path / "bad.gtc"
+    bad.write_bytes(b"gtc\x05" + b"\x00" * 3)
+    r = runner.invoke(app, ["inspect", "--log-json", str(f["idat"]), str(bad)])
+    assert r.exit_code == 3
+    line = next(ln for ln in r.output.splitlines() if ln.startswith('{"schema"'))
+    rec = json.loads(line)
+    assert rec["schema"] == "simurg.log/1" and rec["code"] == "E_GTC_CORRUPT" and rec["level"] == "error"
+
+
+def test_unknown_option_is_usage_error(tmp_path) -> None:  # type: ignore[no-untyped-def]
+    r = runner.invoke(app, ["inspect", "--no-such-option", "x"])
+    assert r.exit_code == 2
+
+
+def test_catalogue_integrity() -> None:
+    for code, d in CATALOGUE.items():
+        assert code == d.code
+        assert code[:2] in ("E_", "W_", "I_")
+        assert d.level in ("error", "warning", "info")
+
+
+def test_messages_doc_is_current() -> None:
+    doc = ROOT / "docs" / "messages.md"
+    assert doc.read_text(encoding="utf-8") == render_markdown(), "regenerate: python -m simurg.core.errors"
+
+
+def test_locale_independent_output(tmp_path) -> None:  # type: ignore[no-untyped-def]  # EDGE-001
+    f = _files(tmp_path)
+    outs = set()
+    for loc in ("C.UTF-8", "tr_TR.UTF-8", "de_DE.UTF-8"):
+        env = {"LC_ALL": loc, "PATH": "/usr/bin:/bin"}
+        r = subprocess.run(
+            [sys.executable, "-m", "simurg.cli.main", "inspect", "--json", str(f["bpm"])],
+            capture_output=True,
+            text=True,
+            env=env,
+            cwd=ROOT,
+        )
+        assert r.returncode == 0, r.stderr
+        outs.add(r.stdout)
+    assert len(outs) == 1