build(server): split Dockerfiles, make ML model paths optional (#120)

Existing Dockerfile (with LLM/STT/TTS model download) is now `Dockerfile.full`.
The new top-level `Dockerfile` builds a slim image without python, without
huggingface_hub, without the model download step. The slim image is the new
default for users who only want the orchestration surface — the directive
folder UI, the mesh/task system, the API.

## Slim Dockerfile
* No python / huggingface_hub / model downloads.
* Same runtime tooling as `k8s/daemon/Dockerfile` (git, gh CLI, ssh, jq,
  curl, ca-certs, libssl3).
* Embeds the daemon binary at /app/daemon-binaries/makima-linux-x86_64
  for the in-server download endpoint.
* PARAKEET_MODEL_DIR / SORTFORMER_MODEL_PATH / CHATTERBOX_MODEL_DIR are
  intentionally NOT set — Listen and Speak return "ML models not
  configured" if a client tries to use them.

## ML model paths now optional
`ServerArgs.parakeet_model_dir`, `parakeet_eou_dir`, `sortformer_model_path`,
`chatterbox_model_dir` are now `Option<String>` (no defaults). The bin
constructor inspects them: if all four are present, configures
`AppState::new`; if all four are absent, uses the new
`AppState::new_slim()` which leaves `model_config = None`. The lazy load
path in `get_ml_models` already returned a clean error for None.

Speak (TTS) was already optional via `model_config.as_ref()` — still works.

Mixed configurations log a warning and degrade to slim mode.

## Ops note
The old `Dockerfile.full` retains the original behaviour for anyone who
needs STT/diarization/TTS in production. CI still builds the daemon image
from `k8s/daemon/Dockerfile` (untouched).

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

1 files changed, 27 insertions, 7 deletions

diff --git a/makima/src/bin/makima.rs b/makima/src/bin/makima.rs
index 338d8f9..a84c581 100644
--- a/makima/src/bin/makima.rs
+++ b/makima/src/bin/makima.rs
@@ -42,13 +42,33 @@ async fn run_server(
     eprintln!("=== Makima Server Starting ===");
     eprintln!("Port: {}", args.port);
 
-    // Create app state
-    let mut app_state = makima::server::state::AppState::new(
-        &args.parakeet_model_dir,
-        &args.parakeet_eou_dir,
-        &args.sortformer_model_path,
-        &args.chatterbox_model_dir,
-    );
+    // Create app state. ML model paths are optional now — when none of
+    // them are supplied (the slim Dockerfile case) Listen and Speak
+    // websocket endpoints respond with "not configured" and everything
+    // else works normally.
+    let mut app_state = match (
+        args.parakeet_model_dir.as_deref(),
+        args.parakeet_eou_dir.as_deref(),
+        args.sortformer_model_path.as_deref(),
+        args.chatterbox_model_dir.as_deref(),
+    ) {
+        (Some(p), Some(eou), Some(sf), Some(cb)) => {
+            eprintln!("ML models configured (lazy load on first use)");
+            makima::server::state::AppState::new(p, eou, sf, cb)
+        }
+        (None, None, None, None) => {
+            eprintln!("ML models NOT configured — Listen/Speak disabled");
+            makima::server::state::AppState::new_slim()
+        }
+        _ => {
+            eprintln!(
+                "WARNING: only some ML model paths provided. Pass all four \
+                 (parakeet/parakeet_eou/sortformer/chatterbox) to enable ML, \
+                 or none to run in slim mode. Continuing in slim mode."
+            );
+            makima::server::state::AppState::new_slim()
+        }
+    };
 
     // Connect to database if URL provided
     if let Some(ref db_url) = args.database_url {