fix: Resolve duplicate columns in Stunner DLS output (#1213)

## Summary
- Stunner parser now recognizes alternate DLS column names used by newer
instrument software (e.g., `Z-Avg Dia. (nm)` vs `Z Ave. Dia (nm)`,
`Rayleigh ratio R (1/km)` vs `(cm^-1)`, `Viscosity at 25°C` vs `20°C`)
- Previously unmapped DLS fields (`Number of Peaks`, `Number of Angles`,
`Angles Measured`, `Intercept`) are now structured as calculated data
documents instead of leaking to custom info
- `number_of_averages` and `Number of acquisitions used` now try both
`(total)` and non-`(total)` column name variants
- `B/S/F/R` (sample role type) added to sample custom info
- Removed 52 N/A placeholder columns from `data system document.custom
information document` that were causing duplicate columns in connector
CSV output

## Context
Follow-up to #1205. Customer (ProFound) reported 17 remaining duplicate
columns after the PkN fix. Root cause was twofold:
1. `CALCULATED_DATA_LOOKUP` did not recognize newer column naming
conventions, so DLS fields fell through `get_unread()` into
`measurement_custom_info`
2. When the parser reads metadata from the first data row (no-header
mode), unread DLS/peak columns leaked into
`data_system_document.custom_information_document` as N/A placeholders —
the CSV conversion lib then produced one column from the placeholder and
one from the actual structured data

## Test plan
- [x] All 34 Stunner/Lunatic tests pass (33 existing + 1 new)
- [x] All 12 Lunatic discovery tests pass
- [x] New anonymized test file (example03) exercises the new PkN column
format with blank, no-peaks, single-peak, and multi-peak measurements
- [x] Verified customer file: 0/85 measurements have
`measurement_custom_info` content
- [x] Verified `data_system_document.custom_information_document` is
empty (was 52 N/A keys)
- [x] Lint clean (black, ruff, mypy)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: nathan-stender

src/allotropy/parsers/unchained_labs_lunatic_stunner/constants.py

@@ -23,7 +23,7 @@
     "The plate data does not contain absorbance measurement for {}."
 )
 
-CALCULATED_DATA_LOOKUP = {
+CALCULATED_DATA_LOOKUP: dict[str, list[dict[str, str | list[str]]]] = {
     "a260": [
         {
             "column": "a260 concentration (ng/ul)",
@@ -102,6 +102,7 @@
         },
         {
             "column": "z ave. dia (nm)",
+            "alt_columns": ["z-avg dia. (nm)"],
             "name": "Z Average Diameter",
             "feature": "dynamic light scattering",
             "unit": "nm",
@@ -114,6 +115,7 @@
         },
         {
             "column": "sd dia (nm)",
+            "alt_columns": ["sd dia. (nm)"],
             "name": "Diameter Standard Deviation",
             "feature": "dynamic light scattering",
             "unit": "nm",
@@ -168,6 +170,7 @@
         },
         {
             "column": "rayleigh ratio r (cm^-1)",
+            "alt_columns": ["rayleigh ratio r (1/km)"],
             "name": "Rayleigh ratio R",
             "feature": "dynamic light scattering",
             "unit": UNITLESS,
@@ -192,6 +195,7 @@
         },
         {
             "column": "viscosity at 20°c (cp)",
+            "alt_columns": ["viscosity at 25°c (cp)"],
             "name": "Viscosity at 20°C (cP)",
             "feature": "dynamic light scattering",
             "unit": UNITLESS,
@@ -204,6 +208,7 @@
         },
         {
             "column": "ri at 20°c",
+            "alt_columns": ["ri at 25°c"],
             "name": "RI at 20°C",
             "feature": "dynamic light scattering",
             "unit": UNITLESS,
@@ -214,5 +219,29 @@
             "feature": "dynamic light scattering",
             "unit": "nm",
         },
+        {
+            "column": "number of peaks",
+            "name": "Number of Peaks",
+            "feature": "dynamic light scattering",
+            "unit": UNITLESS,
+        },
+        {
+            "column": "number of angles",
+            "name": "Number of Angles",
+            "feature": "dynamic light scattering",
+            "unit": UNITLESS,
+        },
+        {
+            "column": "angles measured (°)",
+            "name": "Angles Measured",
+            "feature": "dynamic light scattering",
+            "unit": UNITLESS,
+        },
+        {
+            "column": "intercept",
+            "name": "Intercept",
+            "feature": "dynamic light scattering",
+            "unit": UNITLESS,
+        },
     ],
 }

src/allotropy/parsers/unchained_labs_lunatic_stunner/unchained_labs_lunatic_stunner_calcdocs.py

@@ -262,6 +262,34 @@ def create_calculated_data(
                 view_data=detection_type_view,
                 source_configs=(dynamic_light_scattering_conf,),
             ),
+            CalculatedDataConfig(
+                name="Number of Peaks",
+                value="number of peaks",
+                unit=UNITLESS,
+                view_data=detection_type_view,
+                source_configs=(dynamic_light_scattering_conf,),
+            ),
+            CalculatedDataConfig(
+                name="Number of Angles",
+                value="number of angles",
+                unit=UNITLESS,
+                view_data=detection_type_view,
+                source_configs=(dynamic_light_scattering_conf,),
+            ),
+            CalculatedDataConfig(
+                name="Angles Measured",
+                value="angles measured (°)",
+                unit=UNITLESS,
+                view_data=detection_type_view,
+                source_configs=(dynamic_light_scattering_conf,),
+            ),
+            CalculatedDataConfig(
+                name="Intercept",
+                value="intercept",
+                unit=UNITLESS,
+                view_data=detection_type_view,
+                source_configs=(dynamic_light_scattering_conf,),
+            ),
         ]
     )
 

src/allotropy/parsers/unchained_labs_lunatic_stunner/unchained_labs_lunatic_stunner_structure.py

@@ -244,7 +244,9 @@ def _create_measurement(
         well_plate_identifier=well_plate_data.get(str, "plate id"),
         batch_identifier=well_plate_data.get(str, "sample group"),
         firmware_version=header.get(str, "client version"),
-        number_of_averages=well_plate_data.get(float, "number of acquisitions"),
+        number_of_averages=well_plate_data.get(
+            float, ("number of acquisitions", "number of acquisitions (total)")
+        ),
         integration_time=well_plate_data.get(float, "acquisition time (s)"),
         compartment_temperature=well_plate_data.get(float, "temperature (°c)"),
         sample_custom_info={
@@ -259,13 +261,18 @@ def _create_measurement(
             "molecular weight (kda)": well_plate_data.get(
                 float, "molecular weight (kda)"
             ),
+            "sample role type": well_plate_data.get(str, "b/s/f/r"),
         },
         device_control_custom_info={
             "path length mode": well_plate_data.get(str, "path length mode"),
             "pump": well_plate_data.get(str, "pump"),
             "column": header.get(str, "column") or well_plate_data.get(str, "column"),
             "Number of acquisitions used": well_plate_data.get(
-                str, "number of acquisitions used"
+                str,
+                (
+                    "number of acquisitions used",
+                    "number of acquisitions used (total)",
+                ),
             ),
             "Acquisition filtering": well_plate_data.get(str, "acquisition filtering"),
         },
@@ -434,20 +441,27 @@ def _get_calculated_data_values(
     error_documents: list[ErrorDocument] = []
     for wavelength_column in wavelength_columns:
         for item in CALCULATED_DATA_LOOKUP.get(wavelength_column, []):
-            value, is_na = _get_calculated_value_and_is_na(
-                well_plate_data, item["column"]
-            )
-            # Only create error docs when the cell is the literal "N/A"
+            canonical_column = str(item["column"])
+            alt_columns = item.get("alt_columns", [])
+            columns_to_try = [
+                canonical_column,
+                *(alt_columns if isinstance(alt_columns, list) else []),
+            ]
+            value: float | None = None
+            is_na = False
+            for col in columns_to_try:
+                value, is_na = _get_calculated_value_and_is_na(well_plate_data, col)
+                if value is not None:
+                    break
             if is_na:
                 error_documents.append(
                     ErrorDocument(
                         error=NOT_APPLICABLE,
-                        error_feature=item["name"],
+                        error_feature=str(item["name"]),
                     )
                 )
-            # Skip missing cells entirely; include numeric values
             if value is not None:
-                calculated_data_values[item["column"]] = value
+                calculated_data_values[canonical_column] = value
     return calculated_data_values, error_documents
 
 
@@ -506,6 +520,40 @@ def create_metadata(header: SeriesData, file_path: str) -> Metadata:
                     "background wvl. (nm)",
                     "plate id",
                     "plate position",
+                    "analyte",
+                    "buffer",
+                    "b/s/f/r",
+                    "acquisition filtering",
+                    "number of acquisitions",
+                    "number of acquisitions used",
+                    "number of acquisitions (total)",
+                    "number of acquisitions used (total)",
+                    "acquisition time (s)",
+                    "temperature (°c)",
+                    "molecular weight (kda)",
+                    # DLS calculated data columns
+                    "intercept",
+                    "number of peaks",
+                    "number of angles",
+                    r"^angles measured.*",
+                    r"^z.avg dia.*",
+                    "pdi",
+                    r"^sd dia.*",
+                    r"^diffusion coefficient.*",
+                    r"^rayleigh ratio r.*",
+                    r"^optical contrast constant.*",
+                    r"^viscosity at.*",
+                    r"^ri at.*",
+                    r"^derived intensity.*",
+                    r"^diameter @.*",
+                    r"^kc/r.*",
+                    r"^kd .*",
+                    r"^b22.*",
+                    # Peak columns (pk1, pk2, ..., pkoi, peak 1, peak 2, ...)
+                    r"^pk\d+.*",
+                    r"^pkoi.*",
+                    r"^peak \d+.*",
+                    r"^peak of interest.*",
                     # Strings to skip since these are already captured as measurements/calculated data
                     # Skip absorbance spectrum measurements with a### (10mm)
                     r"^a\d{3} \(10mm\)$",