perf(gaussian): drop the sqrt round-trip from variance-space operations

`Add`, `Sub`, `exclude` and `forget` combined variances by way of standard
deviations: `sigma()` takes a square root, `.powi(2)` squares it away,
`var.sqrt()` takes another, and `from_ms` squares that one back. Three roots
to compute a value that is `1/pi` all along.

They now go through `variance()` and a new `from_mv(mu, var)`, which skip
both conversions. `Sub` is the hot one — `RankDiffFactor::propagate` is
`a - b`, run for every adjacent team pair on every forward and backward
sweep of every EP iteration.

`run_chain` also stopped recomputing each team's weighted performance in the
likelihood loop; the fold is already in `arena.team_prior`, indexed by the
sorted position the loop has in hand. Each `performance()` is itself a
`forget`, so the duplicate cost scaled with players per team.

Measured on this machine, before and after, same fixtures:

    Batch::iteration          23.57us -> 19.31us   (-18%)
    scored_history_60_events   1.071ms -> 983us    (-8%)

The `Gaussian::add`/`sub` microbenchmarks cannot resolve the change: they
sit at ~234ps against a ~218ps floor that `mul`/`div` also hit, so the
harness overhead dominates a single operation.

One golden moved. Two identical competitors drawing must land on their
shared prior mean exactly, by symmetry; the root-free path now returns
25.0 where the reference transcription recorded 24.999999 — that value
rounded to six decimals. Asserting a six-decimal transcription at
epsilon 1e-6 left no headroom, so the expectation is now the exact value.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DnsaJg74eNSva3PJjK2eej

tests/equivalence.rs

@@ -48,15 +48,9 @@ fn game_1v1_draw_golden() {
     )
     .unwrap();
     let p = g.posteriors();
-    // Historical golden from pre-T2 test_1vs1_draw:
-    assert_ulps_eq!(
-        p[0][0],
-        Gaussian::from_ms(24.999999, 6.469480),
-        epsilon = 1e-6
-    );
-    assert_ulps_eq!(
-        p[1][0],
-        Gaussian::from_ms(24.999999, 6.469480),
-        epsilon = 1e-6
-    );
+    // Historical golden from pre-T2 test_1vs1_draw. The mean is 25.0 exactly
+    // by symmetry — two identical competitors drawing cannot move apart — and
+    // the reference's 24.999999 is that value transcribed to six decimals.
+    assert_ulps_eq!(p[0][0], Gaussian::from_ms(25.0, 6.469480), epsilon = 1e-6);
+    assert_ulps_eq!(p[1][0], Gaussian::from_ms(25.0, 6.469480), epsilon = 1e-6);
 }