vLLM priority scheduling for agents

Thread request priority through the API call chain to vLLM's
priority scheduler. Lower value = higher priority, with preemption.

Priority is set per-agent in the .agent header:
- interactive (runner): 0 (default, highest)
- surface-observe: 1 (near-realtime, watches conversation)
- all other agents: 10 (batch, default if not specified)

Requires vLLM started with --scheduling-policy priority.

Co-Authored-By: Proof of Concept <poc@bcachefs.org>

src/subconscious/llm.rs

@@ -22,7 +22,7 @@ pub(crate) fn call_simple(caller: &str, prompt: &str) -> Result<String, String>
 
     let prompts = vec![prompt.to_string()];
     let phases = vec![];
-    super::api::call_api_with_tools_sync(caller, &prompts, &phases, None, &[], None, &log)
+    super::api::call_api_with_tools_sync(caller, &prompts, &phases, None, 10, &[], None, &log)
 }
 
 /// Call a model using an agent definition's configuration (multi-step).
@@ -34,7 +34,7 @@ pub(crate) fn call_for_def_multi(
     bail_fn: Option<&(dyn Fn(usize) -> Result<(), String> + Sync)>,
     log: &(dyn Fn(&str) + Sync),
 ) -> Result<String, String> {
-    super::api::call_api_with_tools_sync(&def.agent, prompts, phases, def.temperature, &def.tools, bail_fn, log)
+    super::api::call_api_with_tools_sync(&def.agent, prompts, phases, def.temperature, def.priority, &def.tools, bail_fn, log)
 }