CLI-1121: Fix sql imports

[danepowell]

Motivation

Drush SQL imports can fail if the SQL file is larger than roughly 64k. For example: acli drush site.env -- sqlc < big-sql-dump.sql

This is due to an upstream Symfony bug: symfony/symfony#21580

Proposed changes

Implement the same workaround as consolidation/site-process#74 and https://github.com/acquia/mc-cs-kafka-consumer/commit/acca146d7736f80add46c9ba5864a0b53758dc71

I'm not confident this is a good workaround. While it fixes this problem, it may introduce others (as evidenced by the failing tests). For instance, acli drush site.env -- sql-dump produces a zero-byte export. Which is weird, because without the redirect it does print to stdout as expected.

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (b0457a7) 91.84% compared to head (8a40257) 91.75%.

❗ Current head 8a40257 differs from pull request most recent head ce0e236. Consider uploading reports for the commit ce0e236 to get more accurate results

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #1581      +/-   ##
============================================
- Coverage     91.84%   91.75%   -0.10%     
- Complexity     1819     1822       +3     
============================================
  Files           124      124              
  Lines          6514     6522       +8     
============================================
+ Hits           5983     5984       +1     
- Misses          531      538       +7     

Files	Coverage Δ
src/Helpers/LocalMachineHelper.php	64.90% <93.75%> (+0.56%)	⬆️

... and 42 files with indirect coverage changes

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

[danepowell]

The latest commit helps but breaks other things, including checklists (in push:artifact, for instance).

[danepowell]

I just realized, it may not be this simple. I think this actually needs to be fixed in Drush since ACLI is calling Drush. We'd only need to "fix" this in ACLI if some other process was calling ACLI and experiencing the bug.

May need to get traction on consolidation/site-process#74

[danepowell]

Now I understand why this appeared to work initially. The replacement error handler was swallowing the pipe error, and the lack of a callback was suppressing the SQL import error. But in fact, nothing was ever fixed here. I think this needs to be fixed upstream in Symfony and/or site-process and/or Drush first.