Perf regression: tls-renegotiation-kestrel, updates

| Scenario | Environment | Date | Old RPS | New RPS | Change | Deviation | StDev | Dependencies |
| -------- | ----------- | ---- | ------- | ------- | ------ | --------- | ----- | ------------ |
| tls-renegotiation-kestrel | Kestrel Linux: TLS Renegotiation 13- Trends cobalt-hosted-lin-azl3 | 07/26/2026 20:22:20 | 1,861 | 1,517 | -100.00 % (-344) :thumbsdown: | -2.98 σ | 115.38 | Changes NameVersionDiff |
| updates [Fixed] | Updates Platform 20- Trends Database cobalt-hosted-lin-azl3-28 | 07/26/2026 23:21:10 | 33,208 | 28,502 | -14.17 % (-4,707) :thumbsdown: | -2.50 σ | 1,881.84 | Changes NameVersionDiff |

[PowerBI Dashboard](https://aka.ms/aspnet/benchmarks)

@sebastienros

The issue names no source file, test, or code entry point. Start with the tls-renegotiation-kestrel results and the linked PowerBI Dashboard, comparing the old and new RPS values and the listed Updates dependency. Done means identifying the cause of the regression and confirming that the benchmark no longer shows it.

Written by the indexing model from the issue text.