test(nwis): parse with raw read_rdb in empty-peaks regression test

The empty-peaks test parsed with _read_rdb, which already runs
format_response(service=None); the real get_discharge_peaks path uses
the raw read_rdb parser followed by format_response(service="peaks").
Switch to read_rdb so the test exercises the actual call path without
the redundant format pass.

Signed-off-by: thodson-usgs <thodson@usgs.gov>

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Author: thodson-usgs

tests/nwis_test.py

@@ -20,6 +20,7 @@
     get_record,
     get_water_use,
     preformat_peaks_response,
+    read_rdb,
 )
 
 START_DATE = "2018-01-24"
@@ -317,7 +318,9 @@ def test_no_peaks_flows_through_format_response(self):
             "# //Response-Status: OK\n"
             "# //Response-Message: No sites found matching all criteria\n"
         )
-        df = _read_rdb(no_peaks_rdb)
+        # Mirror get_discharge_peaks: raw read_rdb, then the peaks-specific
+        # format_response (not _read_rdb, which formats with service=None).
+        df = read_rdb(no_peaks_rdb)
         df = format_response(df, service="peaks")
         assert isinstance(df, pd.DataFrame)
         assert df.empty