Renumber spec 001 decision placeholders to D-040–D-051; record the Danish publication

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

server-py/sidecar.py

@@ -11,7 +11,7 @@ stdio — one JSON object per line, no framing beyond that:
              "audio_path": "/tmp/...", "language": "en" | null,
              "task": "transcribe" | "translate",
              "mode": "file" | "live",                 # optional, default "file"
-             "prompt": "Maxwell, Voices of the Void" | null}  # optional (D-WS2a)
+             "prompt": "Maxwell, Voices of the Void" | null}  # optional (D-048)
   stdout <- {"type": "segment", "id": "1", "start": 0.0, "end": 4.2,
              "text": " Hello world."}                       # 0..n of these
   stdout <- {"type": "done", "id": "1", "cancelled": false}
@@ -26,7 +26,7 @@ container decoding happens here. Requests are processed strictly one at a
 time; a reader thread keeps consuming stdin meanwhile so `cancel` can
 interrupt the segment loop of the request in flight.
 
-Decoding strategy by mode (Decisions.md D-WS5a):
+Decoding strategy by mode (Decisions.md D-047):
   "file"  faster-whisper's BatchedInferencePipeline with its Silero VAD
           (vad_filter=True, VAD_PARAMETERS): silence is dropped, speech is cut
           into <= 30 s chunks and BATCH_SIZE chunks decode per forward pass.
@@ -62,18 +62,18 @@ import sys
 import threading
 
 SAMPLE_RATE = 16000
-BATCH_SIZE = 8  # D-WS5a: VAD chunks decoded per batched forward pass
-# D-WS5a: Silero VAD tuned for file mode. faster-whisper's batched default
+BATCH_SIZE = 8  # D-047: VAD chunks decoded per batched forward pass
+# D-047: Silero VAD tuned for file mode. faster-whisper's batched default
 # (threshold 0.5, 400 ms padding) dropped quiet and overlapped speech —
 # deletions nearly doubled on en_alien; 0.2 / 800 ms brought WER back to the
 # sequential decoder's level. min_silence_duration_ms stays at VadOptions'
 # 2000 ms default (a dict replaces the pipeline's 160 ms preset).
 VAD_PARAMETERS = {"threshold": 0.2, "speech_pad_ms": 800}
 
-# "Names and terms" vocabulary hint (D-WS2a). Twins of normalizeHint /
+# "Names and terms" vocabulary hint (D-048). Twins of normalizeHint /
 # fitHintToBudget in src/lib/hint.ts — change both together; parity is tested
 # in tests/hint-parity.test.ts. Whisper would allow 223 tokens; 120 is the
-# measured cap at which small models are not garbled (D-WS2a).
+# measured cap at which small models are not garbled (D-048).
 PROMPT_TOKEN_BUDGET = 120
 _CONTROL_CHARS = re.compile("[\x00-\x08\x0b\x0c\x0e-\x1f\x7f]")
 
@@ -215,7 +215,7 @@ def main():
             )
             if mode == "live":
                 if prompt_ids:
-                    # D-WS2a: token ids, already cut to the budget by the
+                    # D-048: token ids, already cut to the budget by the
                     # same rule as the browser engine.
                     options["initial_prompt"] = prompt_ids
                 segments, _info = cached_model.transcribe(
@@ -241,7 +241,7 @@ def main():
             # `segments` is a generator: inference happens during iteration,
             # so segments stream out (and cancel can cut in) while later ones
             # are still being decoded — per segment in live mode, per batch of
-            # BATCH_SIZE speech chunks in file mode (D-WS5a).
+            # BATCH_SIZE speech chunks in file mode (D-047).
             for segment in segments:
                 if request_id in cancelled:
                     was_cancelled = True