docs(deepseek): use triple-quoted strings for JSON configs

Switch from escaped inline strings ("{\"method\":\"mtp\"}")
to TOML triple-quoted strings.  The JSON content is much easier to
read and edit:

    --speculative-config, """
{"method":"mtp","num_speculative_tokens":2}""",

TOML multi-line strings strip the leading newline, so the value
passed to vLLM is unchanged.

recipes/deepseek-v4-flash/recipe.toml

@@ -19,12 +19,14 @@ args = [
     "--max-num-batched-tokens", "8192",
     "--gpu-memory-utilization", "0.82",
     "--enable-prefix-caching",
-    "--speculative-config", "{\"method\":\"mtp\",\"num_speculative_tokens\":2}",
+    "--speculative-config", """
+{"method":"mtp","num_speculative_tokens":2}""",
     "--tokenizer-mode", "deepseek_v4",
     "--tool-call-parser", "deepseek_v4",
     "--enable-auto-tool-choice",
     "--reasoning-parser", "deepseek_v4",
-    "--reasoning-config", "{\"reasoning_parser\":\"deepseek_v4\",\"reasoning_start_str\":\"\",\"reasoning_end_str\":\"\"}",
+    "--reasoning-config", """
+{"reasoning_parser":"deepseek_v4","reasoning_start_str":"","reasoning_end_str":""}""",
     "--default-chat-template-kwargs.thinking", "true",
     "--default-chat-template-kwargs.reasoning_effort", "high",
     "--enable-flashinfer-autotune",