fix(anthropic): merge parallel tool results, ordinal stream tool indices, array system content, strict stream decode

Review round 1 fixes (1 Critical + 5 Important) on the anthropic
adapter/translator.

- CRITICAL: consecutive role:"tool" messages each became their own
  Anthropic user turn. Anthropic requires alternating user/assistant
  turns and 400s on two consecutive same-role messages; parallel tool
  calls (one assistant message with N tool_calls followed by N
  role:"tool" messages) are the standard agentic transcript shape.
  anthropicRequestFromOpenAI now accumulates consecutive tool messages
  into one Anthropic user turn carrying all of their tool_result
  blocks (anthropicToolResultBlock replaces anthropicToolResultMessage).
  Added chat_parallel_tool_results fixture.
- Streaming tool_calls[].index passed Anthropic's raw content-block
  index through, so a tool_use block after a text block emitted
  index:1 instead of the required contiguous 0-based ordinal.
  anthropicStreamState now tracks a per-stream tool-call ordinal
  counter plus a content-block-index -> ordinal map, populated on
  content_block_start and consulted on input_json_delta. Added the
  stream_text_then_tool fixture (text block at index 0, tool_use at
  index 1) proving the emitted index is 0.
- A system message with array-form content ([{"type":"text",...}])
  was silently dropped (only a string content was handled). Added
  systemTextFromContentParts: text parts join with "\n"; a non-text
  part returns a *translateError instead of being dropped. Added
  chat_system_array_content (success) and
  chat_system_array_content_error fixtures.
- Malformed JSON on a recognized stream event (message_start,
  content_block_start, content_block_delta, message_delta) was
  swallowed as (nil, nil), which could silently corrupt a client's
  reassembled tool-call arguments. All four now return an error
  wrapping errUpstream; forwardStream already propagates that and
  withholds the terminal [DONE]. The unrecognized-event-type and
  error-event branches are unchanged. Added
  TestAnthropicStreamState_MalformedEventReturnsError.
- providerAdapter.injectAuth's doc comment claimed a universal
  empty-key no-op; updated to state the per-type divergence
  (openai-type: keyless is a no-op; anthropic-type: the constructor
  rejects an empty key, so injectAuth is never called with one).
- Added case-insensitive-Content-Type coverage for both adapters'
  streaming-vs-JSON gate (previously untested): one subtest each in
  provider_openai_test.go / provider_anthropic_test.go using a fake
  upstream that sends "Text/Event-Stream; charset=utf-8", asserting
  the streaming path (not the JSON fallback) is taken.

Gate: go test ./... -count=1 and -race both green; gofmt, go vet,
fieldalignment, golangci-lint (incl. gosec), govulncheck all clean.

provider_openai_test.go

@@ -364,6 +364,49 @@ func TestOpenAIAdapter_ChatCompletion_Streaming_UpstreamIgnoresStreamFallsBackTo
 	}
 }
 
+// TestOpenAIAdapter_ChatCompletion_Streaming_CaseInsensitiveContentType
+// proves the streaming-vs-JSON-fallback gate matches an upstream's
+// Content-Type case-insensitively — review fix: Content-Type values are
+// case-insensitive per RFC 7231, but the gate previously did a
+// case-sensitive substring match against "text/event-stream" and would
+// wrongly fall back to the JSON path for an upstream that sent, e.g.,
+// "Text/Event-Stream". Distinguishing evidence: the streaming path
+// extracts real usage from the SSE chunks and sets Content-Type to the
+// sseWriter's own lowercase "text/event-stream"; the JSON fallback path
+// would copy the upstream's mixed-case header verbatim and fail to parse
+// usage out of raw SSE text.
+func TestOpenAIAdapter_ChatCompletion_Streaming_CaseInsensitiveContentType(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "Text/Event-Stream; charset=utf-8")
+		w.WriteHeader(http.StatusOK)
+		f, ok := w.(http.Flusher)
+		if !ok {
+			t.Fatalf("httptest ResponseWriter does not implement http.Flusher")
+		}
+		for _, frame := range streamFrames {
+			fmt.Fprintf(w, "data: %s\n\n", frame)
+			f.Flush()
+		}
+		fmt.Fprint(w, "data: [DONE]\n\n")
+		f.Flush()
+	}))
+	defer srv.Close()
+
+	a := newOpenAIAdapter("p1", srv.URL, "sk-test")
+	rec := newRecordingResponseWriter()
+
+	u, err := a.chatCompletion(context.Background(), rec, map[string]any{"model": "gpt-5", "stream": true})
+	if err != nil {
+		t.Fatalf("chatCompletion: %v", err)
+	}
+	if u.prompt != 7 || u.completion != 9 {
+		t.Errorf("usage = %+v, want {prompt:7 completion:9} (proves the streaming path, not JSON fallback, was taken)", u)
+	}
+	if got := rec.header.Get("Content-Type"); got != "text/event-stream" {
+		t.Errorf("Content-Type on client response = %q, want %q (sseWriter's own header; JSON fallback would copy the upstream's mixed-case value verbatim)", got, "text/event-stream")
+	}
+}
+
 // TestOpenAIAdapter_ChatCompletion_Upstream429 proves a non-2xx upstream
 // response returns a *providerHTTPError carrying the status and body,
 // with nothing written to the client — the caller decides envelope vs