fix:(embeddings-bedrock) correct extraction of provider from model_name

[luzanikita]

Description

This pull request fixes a bug in the provider extraction logic for Bedrock
embedding models that caused failures when using regionally-scoped model names
(e.g., "eu.cohere.embed-v4:0"). The issue was that the code was naively splitting
model names on . and taking the first part, which would return the region prefix
instead of the actual provider name, resulting in "Provider not supported" errors.

Bug fix:

• Fixed provider extraction to correctly handle both standard 2-part model names
  (provider.model) and regionally-scoped 3-part model names
  (region.provider.model)
• Added a dedicated _get_provider() method with proper logic to extract the
  provider name from different model name formats
• Replaced all instances of self.model_name.split(".")[0] with calls to the new
  _get_provider() method

Provider extraction improvements:

• Added a new _get_provider() method to BedrockEmbedding that correctly
  extracts the provider name from model names:
  • 2-part format: provider.model (e.g., "amazon.titan-embed-text-v1") →
    returns first part ("amazon")
  • 3-part format: region.provider.model (e.g., "eu.cohere.embed-v4:0") →
    returns middle part ("cohere")
  • Raises ValueError for unexpected formats
• Refactored all provider extraction usages in _get_embedding,
  _get_text_embeddings, and _aget_embedding to use the new _get_provider()
  method for consistency and correctness
• Added comprehensive docstring explaining the different model name formats and
  their handling

Testing enhancements:

• Added test_get_provider_two_part_format() to verify standard model names (e.g.,
  "amazon.titan-embed-text-v1", "cohere.embed-english-v3")
• Added test_get_provider_three_part_format() to verify regionally-scoped model
  names (e.g., "us.amazon.titan-embed-text-v1", "eu.cohere.embed-english-v3",
  "global.amazon.titan-embed-text-v2")
• Added test_get_provider_invalid_format() to verify proper error handling for
  invalid model name formats

Code style and import order:

• Reordered imports in base.py for improved readability and consistency
• Minor formatting changes to improve code consistency (line length, string
  formatting)

Root Cause

The original code used self.model_name.split(".")[0] to extract the provider,
which worked for standard model names like "amazon.titan-embed-text-v1" but
failed for regionally-scoped names like "eu.cohere.embed-v4:0" because it would
return "eu" instead of "cohere", causing the error:

{"output": "Error processing request: Provider not supported", "metadata":
{"error": "Provider not supported"}}

Solution

The new _get_provider() method intelligently handles both formats by counting the
number of dot-separated parts and extracting the correct segment based on the
format.

Fixes # (issue)

New Package?

Did I fill in the tool.llamahub section in the pyproject.toml and provide a
detailed README.md for my new integration or package?

• Yes
• No

Version Bump?

Did I bump the version in the pyproject.toml file of the package I am updating?
(Except for the llama-index-core package)

• Yes
• No

Type of Change

Please delete options that are not relevant.

• 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?

Your pull-request will likely not be merged unless it is covered by some form of
impactful unit testing.

• I added new unit tests to cover this change
• I believe this change is already covered by existing unit tests

Suggested Checklist:

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added Google Colab support for the newly added notebooks.
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I ran uv run make format; uv run make lint to appease the lint gods