v0.18.25 : lenient numeric decoding and timestamps truncating

Highlights

• Adds a option to relax the JSON decoders when encountering numerics represented by JSON strings instead of JSON numbers : #1606
• Adds methods on smithy4s.Timestamp to truncate precision to milliseconds or seconds (smithy4s.Timestamp capturing nanosecond-level precision by default) : #1607

Misc fixes

• Generate transform as a final def for cats-tagless friendliness by @kubukoz in #1588
• Don't render parens for case objects in union member nodes by @kubukoz in #1600
• Include field modifiers in collision avoidance by @kubukoz in #1593
• Bump docusaurus-lunr-search from 3.4.0 to 3.5.0 in /modules/website by @dependabot in #1604
• Closes #1539 kebab caseificiation by @dwalend in #1603
• Fix refined newtypes by @kubukoz in #1595
• Avoid collisions in typed node namespaces, add missing uppercase in alt nodes by @kubukoz in #1602
• Introduce the Scala version into the codegen cache namespace by @bpholt in #1590

New Contributors

• @dwalend made their first contribution in #1603

Full Changelog: v0.18.24...v0.18.25