[Host Ir] stream lowering, first milestone

[samnordmann]

This PR only serves as a reference since it got broken down into the following PRs to ease reviewing and merging:

1. [Host irs] alias and preallocated output support #4144
2. [Host Ir] refactor and cleanup lowering and segmentation #4145
3. [Host ir] support for set reduce and binary op #4146
4. [Host irs] Stream lowering of single device fusions #4147

[github-actions]

Description

• Added support for ParallelType::Stream lowering in Host IR for single device fusions.

• Introduced new tests for stream parallelization in test_host_ir_stream_lowering.cpp.

• Enhanced HostIrEvaluator to handle pre-allocated outputs and stream management.

• Updated HostIrLower to support standalone host operations and custom merging logic.

---

Changes walkthrough 📝

	Relevant files
Tests	4 files test_host_ir_stream_lowering.cpp Added comprehensive tests for stream parallelization          +823/-0  test_host_irs.cpp Added tests for pre-allocated outputs and binary operations +290/-4  test_multidevice_host_ir.cpp Disabled StreamParallelType pass temporarily in some tests +10/-0    test_resharding.cpp Updated SegmentCandidateFinderOptions for custom merging  +1/-1
Enhancement	16 files executor.cpp Enhanced HostIrEvaluator for stream management and pre-allocated outputs +207/-106 stream_parallel_type.cpp Implemented StreamParallelType pass for stream parallelization +386/-0  lower.cpp Added logic to handle standalone host operations and custom merging +72/-26  utils.cpp Updated newOutputDomain and newValLike to support reduction axes +19/-8    fusion_segmenter.cpp Updated codeGenSupportedMerge and deriveSchedulerType for custom merging +4/-11    indexing.cpp Updated select to support reduction axes                                  +8/-2      container.cpp Added print method for aliases in HostIrContainer                +5/-0      executor.h Added methods for alias handling and stream management in HostIrEvaluator +39/-1    stream_parallel_type.h Added StreamParallelType pass declaration                                +35/-0    utils.h Updated function signatures for newOutputDomain and newValLike +11/-3    container.h Added methods for alias management and resetting top-level expressions +16/-0    lower.h Added methods for standalone host operations and custom merging logic +7/-0      indexing.h Updated select function signature to support reduction axes +5/-1      executor.h Added method to access HostIrEvaluator                                      +4/-0      fusion_segmenter.h Updated SegmentCandidateFinderOptions for custom merging  +2/-1      internal_nodes.h Added iterDomain method to ForLoop                                              +4/-0
Configuration changes	1 files CMakeLists.txt Added new source and test files                                                    +2/-0
Additional files	1 files test_multidevice_pipeline.cpp +0/-131

---

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

🧪 PR contains tests

⚡ Recommended focus areas for review