feat(hip): add payload-driven shapes and bulk readback

[fhanuman]

Summary

Add a payload-driven shape path for operations whose output dimensions depend on runtime tensor values.

• introduce hip.readback_shape with bufferization, LLVM lowering, and real/mock runtime support for one synchronized vector readback;
• use bulk readback to size Tile outputs from runtime repeats, including partially static results;
• share constant/stamped Pad, Slice, Expand, Range, and Tile shape rules between converter construction and reification;
• preserve Slice’s distinction between physical destination capacity and runtime logical extent;
• use DPS-init shape fallback when payload values cannot be resolved statically;
• consolidate constant extraction used by payload converters and pre-lowering folds.

The readback operation, bufferization model, lowering, runtime ABI, and payload consumers are included in the same layer.

Stack

This is PR 7 of 12 and depends on PR 6:

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

AI assistance

AI tools assisted with the payload-shape refactoring and stack reconstruction. The changes were checked with the authoritative local build, focused payload LIT, and GPU-vs-CPU shape-operation tests.

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	EP inference failed	-	-
GroupQueryAttention_seq256	EP inference failed	-	-
MatMulNBits_o_seq128	EP inference failed	-	-
QMoE_seq128	EP inference failed	-	-

Threshold: 0.01 | Run: 3875 - Commit: 1c67817

[github-actions]

MorphiZen EP Performance Results

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	failed	-	-	-	-
GroupQueryAttention_seq128	failed	-	-	-	-
matmul_down_seq128	failed	-	-	-	-

EPContext Export Performance

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	failed	-	-	-	-

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	-	-	-	0.00
Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	1	5	128	128	-	-	-	0.00

OGA Wheel Smoke (Python benchmark_e2e.py)

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

Run: 3875 - Commit: 1c67817