fix(shape): canonicalize host reshape provenance

[fhanuman]

Summary

Canonicalize host-derived ONNX Reshape targets so descriptor dimensions remain host SSA instead of requiring synchronized payload readbacks.

• analyze supported shape-producing SSA once per function;
• materialize only complete conservative proofs while preserving synchronized fallback;
• harden allowzero, -1, index-width, CFG, and malformed-IR handling.

Stack

This is PR 3 of 12 and depends on PR 2:

1. Constant carrier/externalization
2. Shape-rule foundation
3. Reshape provenance ← this PR
4. MatMul/Gemm
5. Broadcast/Reduction
6. Loop carriers
7. Payload/Readback
8. Conv/Pool
9. Gather/Tensor
10. Attention/Normalization
11. Explicit contract enforcement
12. Refinement hardening

AI assistance

AI tools assisted with the shape-provenance implementation and stack reconstruction. The changes were checked with the authoritative local build, focused LIT, and a strict GPU-vs-CPU Reshape provenance test.

Made with Cursor

[github-actions]

Thanks for opening a PR!

This project follows LLVM's incremental-development and AI-tool-use
guidance. See CONTRIBUTING.md
for the project workflow.

Before requesting review, please check that:

1. The change is focused. Substantial work links the relevant issue
   or design discussion.
2. The PR documents relevant test results and updates affected
   documentation.
3. If AI tools provided substantial assistance, the description
   explains what was assisted and how it was validated, and commit
   trailers identify the tool. The contributor has reviewed and
   understands the result.

Reviewers are assigned through
CODEOWNERS where ownership
is configured.

[github-actions]

L2 Accuracy Results (EP vs CPU)

Model	Combined L2	Total Elems	Skipped NaN/Inf
conv_test_hybrid	4.8668E-07	64	0
GroupQueryAttention_seq256	25.2366	2621440	0
MatMulNBits_o_seq128	259.906	368640	0
QMoE_seq128	34.957	368640	0

Threshold: 0.01 | Run: 3891 - Commit: 1fb8b3e

[github-actions]

MorphiZen EP Performance Results

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	7.52	6.36	359	3	1243
GroupQueryAttention_seq128	4381.27	1.74488	12	6	312
matmul_down_seq128	527.41	2.39	75	3	354

EPContext Export Performance

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	7.58	44.92	356	3	15589

EPContext Import Performance

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	7.56	9.59	361	3	15760

OGA Benchmark Results

Model	Warmup	Reps	Prompt Len	Gen Tokens	TTFT (ms)	TPS	Peak Mem (GB)	GPU Mem (GB)
gpt-oss-20b-webgpu-int4-rtn-block-32	1	5	128	128	146.1	79.9	1.33	13.53
Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	1	5	128	128	205.4	40.1	1.22	6.43

OGA Wheel Smoke (Python benchmark_e2e.py)

Model	TTFT (ms)	TPS
Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	215	39.5

Run: 3891 - Commit: 1fb8b3e