Normalize namespace and name when encoding purl strings

What does this MR do and why?

Normalize namespace and name when encoding purl strings.

This is a spec non-conformance issue, as the spec says the following:

• If the namespace is not empty:

  • Strip the namespace from leading and trailing '/'
  • Split on '/' as segments
  • Apply type-specific normalization to each segment if needed
  • UTF-8-encode each segment if needed in your programming language
  • Percent-encode each segment
  • Join the segments with '/'
  • Append this to the purl
  • Append '/' to the purl
  • Strip the name from leading and trailing '/'
  • Apply type-specific normalization to the name if needed
  • UTF-8-encode the name if needed in your programming language
  • Append the percent-encoded name to the purl

• If the namespace is empty:

  • Apply type-specific normalization to the name if needed
  • UTF-8-encode the name if needed in your programming language
  • Append the percent-encoded name to the purl

Sbom::PackageUrl#to_s is currently only used in tests, so this bug is not user-facing and does not need a changelog entry.

MR acceptance checklist

This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.

• I have evaluated the MR acceptance checklist for this MR.