Make Powershell stop on errors [depends-on: #3616, blocks: #2310, #3636, #3637]

[tautschnig]

We had failing regression tests, but this went unnoticed because the goto-cl
regression tests were successful (which were the last common run in the cmdlet).

• Each commit message has a non-empty body, explaining why the change was made.
• n/a Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• n/a The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[allredj]

✔️
Passed Diffblue compatibility checks (cbmc commit: 10aeffe).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/95900880

[allredj]

✔️
Passed Diffblue compatibility checks (cbmc commit: aeded54).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/95911410

[allredj]

✔️
Passed Diffblue compatibility checks (cbmc commit: 430c6a6).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/95993085

[allredj]

✔️
Passed Diffblue compatibility checks (cbmc commit: 2d9c00b).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/96062013

[chrisr-diffblue]

Apologies if I don't understand the buildspec + windows combination - but why does regression/goto-cl need to be treated any differently from any of the other subdirectories in regression/ ?

[smowton]

I think it's just that goto-cl is not part of the standard make test target set. Perhaps we should make it so under win32, either by checking $OS or by checking if cl.exe appears to be in the path using which.

[chrisr-diffblue]

Fixing up the make test target to understand that on win32 or whatever, extra targets should be added would probably be preferable/less error prone when moving between CI infrastructure, etc.

[tautschnig]

I'll take care of this in a separate PR. We also have some weird interaction between scripts removing dirs and the makefiles at the moment, which should all go away thanks to the -X/-I options in test.pl.