Docs/abot world multi session

[youngmagician114514]

Description

Adds a bilingual technical blog post describing ABot-World multi-session serving, including session state isolation,
compatible batching, multi-GPU worker ownership, migration, backpressure, and fallback behavior.

The article also presents the validated four-H100 trace results for the multi-session serving path.

Motivation

ABot-World multi-session serving was introduced in #36, but the implementation design and measured serving benefits
were not documented in the TeleFuser Technical Blog.

This PR explains how per-session state remains isolated while compatible sessions share model execution, and records
the observed batching, admission, latency, and delivery results.

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)
• Performance improvement
• Code refactoring
• Documentation update
• Other (please describe):

Changes Made

• Add English and Chinese ABot multi-session serving blog posts.
• Document session isolation, compatible batching, process-NCCL workers, migration, backpressure, and fallback behavior.
• Record the four-H100, 30-minute trace results, including a 15.1% reduction in model calls relative to all-B1 execution.
• Add the article to both blog indexes and the MkDocs navigation.

Testing

• Unit tests pass (pytest tests/)
• Manual testing performed
• Benchmarks added/updated (if applicable)

Test commands:

git diff --check

git diff --check passes. The local environment does not currently provide the mkdocs command, so
mkdocs build --strict was not run locally.

## Checklist

- [ ] Code follows the project's coding standards (ruff)
- [ ] Pre-commit hooks pass (pre-commit run --all-files)
- [ ] All tests pass (pytest tests/)
- [ ] New tests added for new functionality
- [x] Documentation updated (README, CLAUDE.md, docstrings)
- [x] Commit messages are clear and descriptive
- [x] PR title follows the convention: [TYPE] Brief description

## Related Issues

Follow-up documentation for #36.

## Additional Notes

This is a documentation-only follow-up to the merged ABot multi-session serving PR. It does not change runtime code,
configuration defaults, or serving behavior.

The blog includes both English and Chinese versions and follows the existing TeleFuser Technical Blog structure.

## GPU Architecture Support

Not applicable; this PR does not add or modify GPU kernels.

- [ ] SM80 (Ampere, Ada Lovelace)
- [ ] SM90 (Hopper H100)
- [ ] SM100+ (Blackwell)

## Performance Impact

This documentation-only PR has no runtime performance impact.

The documented four-H100 trace completed 5,475 successful model calls for 6,452 chunk items. Compatible B2/B3
batching reduced the number of model calls by 977, or 15.1%, compared with all-B1 execution.