Drop the dead lgtm comments.

GitHub's CodeQL doesn't honor the lgtm[...] inline-suppression syntax --
that was the LGTM.com convention, retired in 2022. The two history_id
alerts are now dismissed via the code-scanning API as confirmed false
positives, so the comments do nothing but sit there as cruft. Remove
them; the original false-positive context lives in the dismissal
reason on alerts #407 and #410.

test/evals/seed_staining_quantification_history.py

@@ -158,9 +158,6 @@ def _standalone_main(argv: Optional[list[str]] = None) -> int:
     # signature only names the test type. Cast at this boundary.
     populator = DatasetPopulator(cast(ApiTestInteractor, interactor))
     history_id = seed_demo_history(populator)
-    # lgtm[py/clear-text-logging-sensitive-data] -- history_id is a Galaxy
-    # history id, not a credential. CodeQL flags it because the populator
-    # was constructed with args.galaxy_api_key.
     print(f"Seeded history '{HISTORY_NAME}' at id {history_id}")
     return 0
 

test/integration/test_live_evals.py

@@ -108,10 +108,6 @@ class TestLiveEvals(IntegrationTestCase):
         documented at the top of the file.
         """
         history_id = seed_demo_history(self.dataset_populator)
-        # lgtm[py/clear-text-logging-sensitive-data] -- history_id is a
-        # Galaxy history id, not a credential. CodeQL flags it because the
-        # populator was constructed with self.galaxy_interactor (which holds
-        # an api_key), so any value derived from it inherits the taint.
         log.info("Seeded staining quantification fixture history: %s", history_id)
 
         datasets = [