feat(eval): Go evaluation harness — LinearProbe, Silhouette, EffectiveRank (#4)

internal/eval: three pure-Go diagnostics on frozen embeddings:
  LinearProbe(emb, y, λ) → val_vol_r2 (OOS R², closed-form ridge, Cholesky)
  Silhouette(emb, labels) → mean silhouette (Euclidean, multi-label, errors on <2 classes)
  EffectiveRank(emb) → Roy effective rank (Jacobi eigenvalues → entropy → exp(H))

cmd/eval/main.go: CLI driver reading embeddings.json (exported by train.py with
EXPORT_EMBEDDINGS=1), standardises per-dim, dispatches to -metric flag.
task eval:probe / eval:silhouette / eval:collapse wired in Taskfile.

8/8 tests pass (red-green: perfect clusters, rank-1, full-rank, noise, constant
target, single-label error). Pure stdlib, no external deps.

Closes #4.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Taskfile.yml

@@ -5,16 +5,28 @@ tasks:
     desc: Run templ generate
     cmds: [templ generate]
   build:
-    desc: Build the binary
+    desc: Build all binaries
     deps: [generate]
-    cmds: [go build -o bin/hostexecutor ./cmd/hostexecutor]
+    cmds:
+      - go build -o bin/jepa-fx-risk ./cmd/jepa-fx-risk
+      - go build -o bin/eval ./cmd/eval
   run:
     deps: [build]
-    cmds: [./bin/hostexecutor]
+    cmds: [./bin/jepa-fx-risk]
   test:
     desc: Run all tests
     deps: [generate]
     cmds: [go test ./... -race]
+
+  eval:probe:
+    desc: "Run linear-probe (val_vol_r2) on embeddings from metrics.json"
+    cmds: [./bin/eval -metric probe]
+  eval:silhouette:
+    desc: "Run silhouette on embeddings vs binary HV labels"
+    cmds: [./bin/eval -metric silhouette]
+  eval:collapse:
+    desc: "Run effective-rank collapse diagnostic"
+    cmds: [./bin/eval -metric erank]
   lint:
     cmds: [golangci-lint run ./...]
   check: