feat(providers): explicit prompt cache for DashScope/Qwen (#1127)

* feat(providers): explicit prompt cache for DashScope/Qwen

Extend OpenAI-compat path with Anthropic-style cache_control:ephemeral
inline blocks for Alibaba DashScope endpoints (provider_type=bailian or
URL contains "dashscope"). Verified live: qwen3.6-plus / qwen3.5-plus /
qwen3-coder-plus all return 99.7% cache hit rate on 6K-token prefix,
yielding ~89% token cost reduction on cached prefix (90% Alibaba discount).

- isDashScope(): 3-source detection (URL + providerType + name) handles
  reverse-proxied endpoints; covers both "dashscope" and "bailian"
- buildRequestBody wraps system content with SplitSystemPromptForCache
  (exported from anthropic_request.go) using <!-- GOCLAW_CACHE_BOUNDARY -->
- Tool prefix cache: cache_control on last tool definition, with 4-marker
  budget guard
- Parse cache_creation_input_tokens from prompt_tokens_details into
  Usage.CacheCreationTokens; propagates via existing span metadata writers
- Runtime escape hatch: GOCLAW_DISABLE_DASHSCOPE_CACHE=true
- Live integration smoke test (build tag integration, env-gated)

* chore(providers): polish per PR #1127 review

- Use strings.Repeat instead of custom repeat() helper in smoke test
- Clarify BuildRequestBodyForTest is test-only, not public API

* feat(providers): enable thinking for Qwen 3.7/3.6, observe cache in smoke test

Qwen3.7-plus and qwen3.6-plus support deep thinking but were missing from
dashscopeThinkingModels, so enable_thinking/thinking_budget was silently
skipped. Add both to the whitelist and test.

Add qwen3.7-plus to the cache smoke test as cache-optional: Alibaba's
context-cache doc does not yet list 3.7/3.6 for explicit cache, and the
cache wrap is a safe no-op when unsupported, so a no-cache result is logged
rather than failed to avoid a flaky live assertion.

Claude-Session: https://claude.ai/code/session_01X3jkrc7N3ar8ZGzUJWyNS5

* fix(providers): preserve DashScope detection for proxy routes

internal/providers/anthropic_cache_split_test.go

@@ -8,7 +8,7 @@ import (
 // the cache boundary marker is split into 2 blocks: stable (cached) + dynamic.
 func TestAnthropicSystemBlocksSplit(t *testing.T) {
 	prompt := "stable content\n" + CacheBoundaryMarker + "\ndynamic content"
-	blocks := splitSystemPromptForCache(prompt)
+	blocks := SplitSystemPromptForCache(prompt)
 	if len(blocks) != 2 {
 		t.Fatalf("expected 2 blocks, got %d", len(blocks))
 	}
@@ -30,7 +30,7 @@ func TestAnthropicSystemBlocksSplit(t *testing.T) {
 // marker → single block with cache_control.
 func TestAnthropicSingleBlockFallback(t *testing.T) {
 	prompt := "no boundary here"
-	blocks := splitSystemPromptForCache(prompt)
+	blocks := SplitSystemPromptForCache(prompt)
 	if len(blocks) != 1 {
 		t.Fatalf("expected 1 block, got %d", len(blocks))
 	}
@@ -46,7 +46,7 @@ func TestAnthropicSingleBlockFallback(t *testing.T) {
 // after the boundary produces only 1 block (no empty block appended).
 func TestAnthropicEmptyDynamic(t *testing.T) {
 	prompt := "stable only\n" + CacheBoundaryMarker + "\n"
-	blocks := splitSystemPromptForCache(prompt)
+	blocks := SplitSystemPromptForCache(prompt)
 	if len(blocks) != 1 {
 		t.Fatalf("expected 1 block for empty dynamic, got %d", len(blocks))
 	}
@@ -59,7 +59,7 @@ func TestAnthropicEmptyDynamic(t *testing.T) {
 // (empty stable section) still produces valid blocks without empty text.
 func TestAnthropicEmptyStable(t *testing.T) {
 	prompt := CacheBoundaryMarker + "\ndynamic only"
-	blocks := splitSystemPromptForCache(prompt)
+	blocks := SplitSystemPromptForCache(prompt)
 	// Stable is empty string after TrimSpace — should still produce a block
 	// (Anthropic API handles empty text blocks gracefully).
 	if len(blocks) < 1 {