test(wtr): run integration tests in native and synthetic shadow modes @W-19098130

[wjhsf]

Details

1. Integration tests run by default in synthetic shadow, while hydration tests run by default in native shadow. There were quasi-duplicate env vars, DISABLE_SYNETHETIC (integration) and ENABLE_SYNTHETIC_SHADOW_IN_HYDRATION (hydration). They weren't set up quite right, so I fixed them. And then I consolidated them into a single env var, SHADOW_MODE_OVERRIDE. This required a minor shuffling around of options so that NATIVE_SHADOW could be set with the correct default value in each test suite.
2. I updated a handful of tests, so now all tests pass in both native and synthetic shadow modes.
3. I updated CI so that integration and hydration tests run in native and synthetic shadow modes, and CI will fail if the other tests fail.
4. I added all env vars modes to CI. The ones not specified above are just informational, if they fail then CI will still pass.
   a. This makes CI take ~35 minutes, but that's still shorter than nucleus, so it shouldn't be too much of a slowdown. I'm also actively working on this, and I will eventually parallelize test runs so that we stay close to the current ~15 minutes.

Does this pull request introduce a breaking change?

• 😮‍💨 No, it does not introduce a breaking change.
• 💔 Yes, it does introduce a breaking change.

Does this pull request introduce an observable change?

• 🤞 No, it does not introduce an observable change.
• 🔬 Yes, it does include an observable change.

GUS work item

[wjhsf]

Oops! We were using the wrong object, so none of the tests were actually executing. Fortunately, all the tests are still passing with this fixed. I added the else clause below to prevent this accident from happening again.