Typing analysis

Note: Ignored files are excluded from the next sections.

steep:ignore comments

This PR introduces 1 steep:ignore comment, and clears 1 steep:ignore comment.

lib/datadog/tracing/distributed/baggage.rb:216

lib/datadog/tracing/distributed/baggage.rb:180

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 97.21% (-0.01%)

Benchmarks

Benchmark execution time: 2026-05-05 23:10:00

Comparing candidate commit 805b8be in PR branch fix-baggage with baseline commit 483848f in branch master.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 45 metrics, 1 unstable metrics.

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

The byteslice(0, MAX+1) + chop choice for the truncation case is clean — chop walks back to the last valid character boundary, so 1, 2, or 3 trailing partial bytes get dropped in a single step. The simpler byteslice(0, MAX) would have left invalid bytes in some inputs.

A follow-up on the same UTF-8 dimension, after going through the rest of the parser. Two pre-existing raise paths still reach the caller of Propagation#extract, since propagate_baggage runs outside the rescue block at propagation.rb:146:

1. UTF-8-tagged input with invalid byte sequences → String#split and String#strip raise ArgumentError: invalid byte sequence in UTF-8.
2. Malformed percent-encoding (%XX, lone %) → URI.decode_www_form_component raises ArgumentError: invalid %-encoding.

Both pre-existing, not introduced here. Flagging because this PR is already focused on UTF-8 in this method, and the new byte cap creates a small envelope where a 9KB invalid-UTF-8 header gets handled cleanly via byteslice+chop+remove_last_entry → {}, while the same bytes in a 100-byte header still raise.

Opened #5689 as a self-contained companion fix to make the recommendation easy to evaluate alongside this PR. It adds a valid_encoding? guard at the top of parse_baggage_header and a rescue ArgumentError around the two URI.decode_www_form_component calls — both route into the existing malformed-telemetry path. ASCII-8BIT input (Rack's typical encoding) is unaffected. Three new specs lock the contract; two of them fail without the production change.

The two PRs touch parse_baggage_header at different lines and merge cleanly in either order — no dependency.