Remove digest artifacts from sequence

[swalker2m]

The ODB API will be updated to provide execution / digest and execution / config queries. The former (digest) returns summary information for which it is required to examine the entire sequence (a fold over all the atoms). The later (config) returns the familiar nextAtom, possibleFuture, and hasMore for each of the acquisition and science sequences. This enables the digest information to be neatly cached while the expanded config can be quickly computed upon request thanks to the tight futureLimit that is imposed upon the sequences.

This PR enables the pending ODB update by moving the atomCount (and setupTime) to the digest.

[codecov-commenter]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.13% 🎉

Comparison is base (c802b8a) 78.61% compared to head (6b483c3) 78.74%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #769      +/-   ##
==========================================
+ Coverage   78.61%   78.74%   +0.13%     
==========================================
  Files         112      112              
  Lines        1487     1487              
  Branches        4        3       -1     
==========================================
+ Hits         1169     1171       +2     
+ Misses        318      316       -2     

Files Changed	Coverage Δ
...a/core/model/sequence/arb/ArbExecutionConfig.scala	87.50% <ø> (-1.39%)	⬇️
...core/model/sequence/arb/ArbExecutionSequence.scala	75.00% <100.00%> (-6.82%)	⬇️
...ma/core/model/sequence/arb/ArbSequenceDigest.scala	75.00% <100.00%> (+3.57%)	⬆️

... and 6 files with indirect coverage changes

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

[rpiaggio]

So the digest is now parallel to execution instead of embedded?

[rpiaggio]

LGTM

[swalker2m]

So the digest is now parallel to execution instead of embedded?

You can ask for both in the same query (and both are now embedded in execution as siblings) but it facilitates the implementation to separate that which needs to read the entire sequence from that which needs only the first few steps.

query {
  observation(observationId: "o-100") {
    execution {
      digest { ... }
      config { ... }
    }
  }
}