agent: bundle sampling fields as SamplingParams on AgentState

Collapse the split `temperature` / `top_p` / `top_k` fields on
AgentState into a single `sampling: SamplingParams` struct, mirroring
how the wire-level fields flow into the Generate RPC. Adds
`max_tokens` to SamplingParams so it's actually plumbed end to end
(previously the client had a hardcoded 4096 fallback inside
`run_session_generate`).

AgentState construction sites now set `sampling: SamplingParams { ...
max_tokens: 4096 }` as the default. The assignment sites in
oneshot.rs / subconscious.rs / unconscious.rs switch from
`st.temperature = X` to `st.sampling.temperature = X`.

`stream_session_mm` takes `SamplingParams` directly; the
`sampling_max_tokens()` helper goes away. `pb::GenerateRequest` is
populated with `sampling.max_tokens` (and the other fields) in
`run_session_generate`. SamplingParams is `pub` so it can be
embedded in the public AgentState without a visibility warning.

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

src/subconscious/generate.rs

@@ -43,6 +43,7 @@ where F: FnMut(&AstNode) -> bool,
         temperature: 0.6,
         top_p: 0.95,
         top_k: 20,
+        max_tokens: 4096,
     };
 
     // Ephemeral per-call session — opens on first touch, drops when