[LPD] [Correct] Ignore missing sources when finding memory cell patterns

Description

This came up with p6936 r117; for some reason some LPD modules were entirely missing from the run (no files created), and then the notebook fails in this function before finding constants, whereas it should be able to still correct the data that is there.

How Has This Been Tested?

Interactively in the LPD correction notebook.

Types of changes

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• My code follows the code style of this project.

Reviewers

@ahmedk @schmidtp

src/cal_tools/lpdlib.py

@@ -4,7 +4,7 @@ from typing import Optional
 from warnings import warn
 
 import numpy as np
-from extra_data import RunDirectory
+from extra_data import RunDirectory, SourceNameError
 
 
 def get_mem_cell_pattern(run, sources) -> np.ndarray:
@@ -15,7 +15,11 @@ def get_mem_cell_pattern(run, sources) -> np.ndarray:
     """
     patterns = []
     for source in sources:
-        cell_id_data = run[source, 'image.cellId'].drop_empty_trains()
+        try:
+            cell_id_data = run[source, 'image.cellId'].drop_empty_trains()
+        except SourceNameError:
+            continue  # Source missing
+
         if len(cell_id_data.train_ids) == 0:
             continue  # No data for this module
         cell_ids = cell_id_data[0].ndarray().flatten()