copy reference pixels during dqinit

[braingram]

dqinit adds data views to the resulting datamodel which when saved triggers asdf to save the base array resulting in large file sizes. Similar to #1098 this PR adds copies of those arrays to the output datamodel to avoid larger-than-needed file sizes.

Without this PR running the elp benchmarks (each step in `elp) with a 330Mb input file produces files with sizes as follows:

step	output file size
dqinit	1.5 G
saturation	1.5 G
refpix	2.2 G
linearity	2.2 G
darkcurrent	2.2 G
rampfit	332 M
assignwcs	332 M
flatfield	395 M
photom	395 M
sourcedetection	395 M

Note the jump to 2.2 G for the refpix output.

With this PR the file sizes are:

step	output file size
dqinit	1.5 G
saturation	1.5 G
refpix	1.5 G
linearity	1.5 G
darkcurrent	1.5 G
rampfit	332 M
assignwcs	332 M
flatfield	395 M
photom	395 M
sourcedetection	395 M

Note that the jump in file size is not seen until later in the pipeline because the base array saved for the views in dqinit is shared until later when the base array is replaced.

Regtests run with no errors: https://plwishmaster.stsci.edu:8081/job/RT/job/Roman-Developers-Pull-Requests/623/

Checklist

• added entry in CHANGES.rst under the corresponding subsection
• updated relevant tests
• updated relevant documentation
• updated relevant milestone(s)
• added relevant label(s)
• ran regression tests, post a link to the Jenkins job below. How to run regression tests on a PR

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 76.73%. Comparing base (1cacc96) to head (046faae).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1121   +/-   ##
=======================================
  Coverage   76.73%   76.73%           
=======================================
  Files         105      105           
  Lines        7054     7054           
=======================================
  Hits         5413     5413           
  Misses       1641     1641           

Flag	Coverage Δ		*Carryforward flag
nightly	62.73% <ø> (-0.09%)	⬇️	Carriedforward from 1cacc96

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[schlafly]

LGTM, thanks!