Fix incorrect AttributeError for invalid AnyValue types (#4821)

[tejaspavanb]

Description

Fixes #4821

Summary

On Python 3.9, passing an invalid AnyValue type could raise an unexpected AttributeError
while constructing the error message, due to certain typing types missing a __name__
attribute.

Motivation and Context

This change ensures that error messages are constructed safely by resolving type names
using a fallback mechanism. It restores the expected TypeError behavior without altering
existing functionality. The fix follows the approach suggested in the issue discussion
and adds a regression test to cover this edge case.

---

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

---

How Has This Been Tested?

• Unit tests

Test details

• Ran python -m pytest opentelemetry-api/tests/attributes
• Added a regression test to verify that a TypeError is raised only when stringification of an invalid value fails
• Verified all existing attribute tests pass on Windows with Python 3.12

---

Does This PR Require a Contrib Repo Change?

• Yes. – Link to PR:
• No.

---

Checklist

• Followed the style guidelines of this project
• Changelogs have been updated (not required for this fix)
• Unit tests have been added
• Documentation has been updated (not required)

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: tejaspavanb / name: Tejas Pavan B (e540604, ff91815)

[xrmx]

Given that 3.9 is deprecated since november and since the report of the issue we've already changed the code to reduce greatly the time this code is called in practice if we really want to merge this, we need to add a comment with a TODO for go back to plain __name__ after we drop 3.9 support in May.

[tejaspavanb]

Thanks for the clarification. I’ve added a TODO comment near the workaround noting that it should be removed and reverted once Python 3.9 support is dropped. Please let me know if the wording looks good.