completion() for fireworks

[dineshyv]

What does this PR do?

implements completetion() API for fireworks

Feature/Issue validation/testing/test plan

 PROVIDER_ID=test-fireworks MODEL_IDS="Llama3.1-8B-Instruct" PROVIDER_CONFIG=../configs/provider_config_example.yaml with-proxy pytest -s tests/inference/test_inference.py --tb=short --disable-warnings -rs
/home/dineshyv/.conda/envs/stack/lib/python3.10/site-packages/pytest_asyncio/plugin.py:208: PytestDeprecationWarning: The configuration option "asyncio_default_fixture_loop_scope" is unset.
The event loop scope for asynchronous fixtures will default to the fixture caching scope. Future versions of pytest-asyncio will default the loop scope for asynchronous fixtures to function scope. Set the default fixture loop scope explicitly in order to avoid unexpected behavior in the future. Valid fixture loop scopes are: "function", "class", "module", "package", "session"

  warnings.warn(PytestDeprecationWarning(_DEFAULT_FIXTURE_LOOP_SCOPE_UNSET))
================================================================================= test session starts =================================================================================
platform linux -- Python 3.10.15, pytest-8.3.3, pluggy-1.5.0
rootdir: /home/dineshyv/local/llama-stack
configfile: pyproject.toml
plugins: asyncio-0.24.0, anyio-4.6.2.post1
asyncio: mode=strict, default_loop_scope=None
collected 8 items

tests/inference/test_inference.py Resolved 4 providers
 inner-inference => test-fireworks
 models => __routing_table__
 inference => __autorouted__
 inspect => __builtin__

........

=========================================================================== 8 passed, 21 warnings in 9.24s ============================================================================

[dineshyv]

@ashwinb, this is needed since the number of chunks each provider returns is not consistent and does not depend on the max tokens.

[ashwinb]

sweet