Add dataset to store list of software that produced a file (#578)

* add nwbfile dataset for provenance information

* update software version dataset name and doc

* update release notes

core/nwb.file.yaml

@@ -241,6 +241,18 @@ groups:
       - name: file_name
         dtype: text
         doc: Name of script file.
+    - name: was_generated_by
+      dtype: text
+      doc: Name and version of software package(s) used to generate data contained in
+       this NWB File. For each software package or library, include the name of the
+       software as the first value and the version as the second value.
+      dims:
+      - num_sources
+      - name, version
+      shape:
+      - null
+      - 2
+      quantity: '?'
     - name: stimulus
       dtype: text
       doc: Notes about stimuli, such as how and where they were presented.

docs/format/source/format_release_notes.rst

@@ -12,7 +12,7 @@ Minor changes
   ``waveform_sd`` columns ragged. This allows for a different number of waveform means/SDs per unit which is useful
   when each unit is associated with a different number of electrodes and there is a waveform mean/SD for each
   electrode and unit. (#576)
-
+- Added optional ``was_generated_by`` attribute to ``NWBFile`` to store provenance information (#578)
 
 2.7.0 (February 7, 2024)
 ------------------------