Use new fileaccess to build execution state

[wild-endeavor]

Tracking issue

COR-982

Why are the changes needed?

In cases where the python interpreter is not shut down between executions, we need to make sure we don't accidentally pick up an old errors.pb file.

What changes were proposed in this pull request?

When starting a new execution, we run the setup_execution context manager. One of the things it does is create a new execution state object, which creates a new engine dir based on the working dir. This working dir is based on the file access provider local_sandbox_dir. That local sandbox dir is created anew upon every invocation, but the new engine dir was happening before it was getting used. This PR just picks up that new workdir first, before building the execution state object.

How was this patch tested?

locally.

Setup process

Screenshots

Check all the applicable boxes

• I updated the documentation accordingly.
• All new and existing tests passed.
• All commits are signed-off.

Related PRs

Docs link

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 74.10%. Comparing base (1a01368) to head (d5b59da).
Report is 3 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2452      +/-   ##
==========================================
+ Coverage   72.22%   74.10%   +1.87%     
==========================================
  Files         182      242      +60     
  Lines       18475    21029    +2554     
  Branches     3608     3609       +1     
==========================================
+ Hits        13344    15583    +2239     
- Misses       4499     4813     +314     
- Partials      632      633       +1     

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