Unstable TestTokenAuthority.tokens_are_revoked_in_time test

[Saviq]

Judging by the test name, it relies on timing things, which in this case didn't work out.

The failure comes from a riscv64 build, running on QEMU, so has the potential for being our slowest.

292/325 Test #295: mir_unit_tests.TestTokenAuthority.* ..............................................................................................***Failed   13.20 sec
Running main() from main.cpp
Note: Google Test filter = TestTokenAuthority.*:-:--
[==========] Running 6 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 6 tests from TestTokenAuthority
[ RUN ] TestTokenAuthority.can_issue_tokens
[ OK ] TestTokenAuthority.can_issue_tokens (33 ms)
[ RUN ] TestTokenAuthority.tokens_are_revoked_in_time
./tests/unit-tests/shell/test_token_authority.cpp:84: Failure
Expected: (delta_ms - revocation_time) < (epsilon), actual: 14ms vs 10ms

[ FAILED ] TestTokenAuthority.tokens_are_revoked_in_time (3043 ms)
[ RUN ] TestTokenAuthority.manual_revocation_works_correctly
[ OK ] TestTokenAuthority.manual_revocation_works_correctly (32 ms)
[ RUN ] TestTokenAuthority.manual_revocation_doesnt_revoke_other_tokens
[ OK ] TestTokenAuthority.manual_revocation_doesnt_revoke_other_tokens (312 ms)
[ RUN ] TestTokenAuthority.revoke_all_actually_revokes_all
[ OK ] TestTokenAuthority.revoke_all_actually_revokes_all (19 ms)
[ RUN ] TestTokenAuthority.bogus_token_is_actually_bogus
[ OK ] TestTokenAuthority.bogus_token_is_actually_bogus (7 ms)
[----------] 6 tests from TestTokenAuthority (3463 ms total)

[----------] Global test environment tear-down
[==========] 6 tests from 1 test suite ran. (3470 ms total)
[ PASSED ] 5 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] TestTokenAuthority.tokens_are_revoked_in_time

1 FAILED TEST

[Saviq]

FWIW to have a reliable test here, we'd need to not rely on actual time, but rather mock and control the time source.