Add Part 9: GPU-Accelerated Semantic Caching with cuVS CAGRA#149

#### Description External contributor provided change for Triton Inference Server Tutorial repository<br>[https://github.com/triton-inference-server/tutorials/pull/149](<https://github.com/triton-inference-server/tutorials/pull/149>) #### Reproduction steps #### Acceptance criteria - [ ] Confirmed/rejected change - [ ] Triton contribution agreement signed by author/user - [ ] Add job to CI if required