Merge t16-grad-accum into main

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

# Conflicts:
#	README.md
#	docs/evolution.md

crates/xtrain-distributed/src/bin/train_ddp.rs

@@ -74,6 +74,10 @@ fn main() {
     // Optimization knobs (mirror bin/train).
     let steps: usize = flag(&args, "--steps", 100);
     let batch: usize = flag(&args, "--batch", 16);
+    // Micro-batch gradient accumulation (Phase T16): effective global batch =
+    // accum_steps × batch, all-reducing only at the accumulation boundary. Default
+    // 1 = no accumulation (bit-identical to the pre-T16 DDP path).
+    let accum_steps: usize = flag(&args, "--accum-steps", 1).max(1);
     let seq_len: usize = flag(&args, "--seq", 64);
     let max_lr: f32 = flag(&args, "--max-lr", 3e-3);
     let min_lr: f32 = flag(&args, "--min-lr", max_lr * 0.1);
@@ -150,6 +154,7 @@ fn main() {
     let dcfg = DdpConfig {
         seq_len,
         batch_size: batch,
+        accum_steps,
         steps,
         schedule: LrSchedule {
             max_lr,
@@ -167,8 +172,9 @@ fn main() {
     };
 
     println!(
-        "training: {steps} steps, seq {seq_len}, global batch {batch}, lr {max_lr:.1e}→{min_lr:.1e}, \
-         eval every {eval_every}"
+        "training: {steps} steps, seq {seq_len}, global batch {batch} × accum {accum_steps} = \
+         effective global batch {}, lr {max_lr:.1e}→{min_lr:.1e}, eval every {eval_every}",
+        batch * accum_steps
     );
 
     if bf16 {