Apply uv tool for dependency version control

[chensuyue]

Description

UV provides a drop-in replacement for common pip, pip-tools commands, deps installation experience a 10-100x speedup with the uv pip interface. https://docs.astral.sh/uv/.

• Purpose

1. Accelerate pip install with uv.
   For examples: docker build -t opea/build:test -f comps/retrievers/src/Dockerfile .
   uv install Vs. pip install is 30.1s Vs.173.6s.

# pip install logs
 => [5/7] RUN pip install --no-cache-dir --upgrade pip setuptools uv &&     if [ cpu = "cpu" ]; then       uv pip install --system --no-cache-dir torch torchvision --index-url https://download.pytorch.org/whl/cpu;       uv pip ins  30.1s
# uv install logs
 => [5/7] RUN pip install --no-cache-dir --upgrade pip setuptools &&     if [ cpu = "cpu" ]; then         PIP_EXTRA_INDEX_URL="--extra-index-url https://download.pytorch.org/whl/cpu";     else         PIP_EXTRA_INDEX_URL="";       173.6s

2. Freeze deps version to avoid issues involved by regular dependency version update.
   For example, transformers, langchain, huggingface_hub updated with high frequency.

• Steps

1. Rename current requirements.txt to requirements.in as dependency template, including the first level dependencies.
2. Run uv pip compile generate requirements.txt or requirements-cpu.txt and requirements-gpu.txt for different HW ENV. Full CMD is $ uv pip compile requirements.in --universal -o requirements.txt --upgrade, requirements.in works as a dependency template, --universal means cross platform, --upgrade means overwrite requirements.txt with latest version.
   CPU version CMD is $ uv pip compile --extra-index-url https://download.pytorch.org/whl/cpu --index-strategy unsafe-best-match requirements.in --universal -o requirements-cpu.txt --upgrade
3. Update Dockerfile, use uv to install requirements.txt.

• TODO

1. Enable a CI/CD workflow to upgrade the deps version regularly.
2. Remove some hardcode version limit in requirements.in and test with CI. Module dependency version differences #1775
3. Add Readme of how to generate requirements.txt.

Issues

List the issue or RFC link this PR is working on. If there is no such link, please mark it as n/a.

Type of change

List the type of change like below. Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds new functionality)
• Breaking change (fix or feature that would break existing design and interface)
• Others (enhancement, documentation, validation, etc.)

Dependencies

List the newly introduced 3rd party dependency if exists.

Tests

Describe the tests that you ran to verify your changes.

[lkk12014402]

LGTM

[chensuyue]

image2image_on_intel_hpu
llms_text-generation_service_vllm_on_intel_hpu
Those 2 tests also failed with main branch, so it's known issue, will file ticket to trace. https://github.com/opea-project/GenAIComps/actions/runs/15813485072/job/44568234934?pr=1809

[chensuyue]

llms_doc-summarization_vllm_on_intel_hpu
llms_faq-generation_vllm
Those 2 tests failed with known issue, test got stuck. #1725

[chensuyue]

The 6 issues found in the last test round have been addressed, 4 known issues, 2 fixed in #1812, so this PR is ready for merge. https://github.com/opea-project/GenAIComps/actions/runs/15818449179

[eero-t]

Why this drops pytorch CPU version override, don't indirect deps bring it in any more?

[chensuyue]

This docker image does not include torch in its dependency, so I remove this --extra-index-url.