Bug: Confusing durations during extended runs

### Description of the issue Output I get during multi-hour test runs: ``` [...] Finished `release` profile [optimized + debuginfo] target(s) in 1m 16s ──────────── Nextest run ID 4ef52442-baeb-4bae-9884-eb5cd6ef4303 with nextest profile: extended Starting 8 tests across 8 binaries (1 test skipped) Running [ 01:37:37] 0/8: 8 running, 0 passed, 1 skipped Running [ 01:37:38] 0/8: 8 running, 0 passed, 1 skipped Running [ 01:37:40] 0/8: 8 running, 0 passed, 1 skipped Running [ 01:37:46] 0/8: 8 running, 0 passed, 1 skipped Running [ 05:13:30] 0/8: 8 running, 0 passed, 1 skipped [05:313:30] omq-tokio::omq_soak_multi_socket soak_multi_socket [05:313:30] omq-tokio::omq_soak_hwm_reconnect soak_hwm_reconnect_block [05:313:30] omq-tokio::omq_soak_large_throughput soak_large_message_throughput [05:313:30] omq-tokio::omq_soak_mechanism_reconnect soak_curve_reconnect [05:313:30] omq-tokio::omq_soak_reconnect_all_types soak_reconnect_all_types [05:313:30] omq-tokio::omq_soak_pub_sub_realworld_churn_tcp soak_pub_sub_realworld_churn_tcp [05:313:30] omq-tokio::omq_soak_router_dealer_churn soak_router_dealer_churn [05:313:30] omq-tokio::omq_soak_reconnect_storm soak_reconnect_storm ``` The durations like `[05:313:30]` are confusing to me. I'm assuming that's a bug. ### Expected outcome Minutes (middle integer) in the range of `0..=59` ### Actual result Minutes (middle integer) outside of range `0..=59`, like `313` ### Nextest version ```text cargo-nextest 0.9.140 (a9fef2964 2026-07-05) release: 0.9.140 commit-hash: a9fef2964e34f64ed4fceeee7c0c3559ce560920 commit-date: 2026-07-05 host: x86_64-unknown-linux-gnu ``` ### Additional context _No response_