Setup QuantEcon-wasm library

[kp992]

Setup QuantEcon-wasm library. This PR mirrors the original QuantEcon.py library and removes numba dependency for WASM environments.

[kp992]

Locally all the tests are passing:

% pytest -n auto
============================================================================ test session starts =============================================================================
platform darwin -- Python 3.10.15, pytest-8.1.0, pluggy-1.4.0
rootdir: /Users/kpl/repos/QuantEcon.py-wasm
configfile: pytest.ini
plugins: anyio-4.0.0, xdist-3.6.1
8 workers [532 items]   
....................................................................................................................................................................... [ 31%]
............................x......................................................................................................................................... [ 62%]
...................................................................................................................................................................... [ 93%]
.................................                                                                                                                                      [100%]
============================================================================== warnings summary ==============================================================================
quantecon_wasm/markov/tests/test_ddp.py::TestFiniteHorizon::test_backward_induction
quantecon_wasm/markov/tests/test_ddp.py::test_ddp_beta_1_not_implemented_error
  /Users/kpl/repos/QuantEcon.py-wasm/quantecon_wasm/markov/ddp.py:430: UserWarning: infinite horizon solution methods are disabled with beta=1
    warnings.warn(msg)

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
================================================================ 531 passed, 1 xfailed, 2 warnings in 57.99s =================================================================

One of the tests is failing and is known to me. It's marked with xfail.

[mmcky]

@kp992 I will dive into this error on Thur/Fri this week.

[mmcky]

@kp992 an fail is an expected fail so that isn't causing the failure

Looks like tolerances are different

=========================== short test summary info ============================
FAILED quantecon_wasm/tests/test_quad.py::TestQnwbeta::test_qnwbeta_nodes_3d - AssertionError: 
Not equal to tolerance rtol=1e-07, atol=0

Mismatched elements: 459 / 945 (48.6%)
Max absolute difference among violations: 0.92125805
Max relative difference among violations: 2.55069065
 ACTUAL: array([[0.022479, 0.236368, 0.022315],
       [0.26579 , 0.236368, 0.022315],
       [0.943737, 0.236368, 0.022315],...
 DESIRED: array([[0.022479, 0.236368, 0.022315],
       [0.114679, 0.236368, 0.022315],
       [0.26579 , 0.236368, 0.022315],...
FAILED quantecon_wasm/tests/test_quad.py::TestQnwbeta::test_qnwbeta_weights_3d - AssertionError: 
Not equal to tolerance rtol=1e-07, atol=0

Mismatched elements: 297 / 315 (94.3%)
Max absolute difference among violations: 0.02843322
Max relative difference among violations: 9.73330951
 ACTUAL: array([7.406938e-05, 1.685941e-04, 6.900889e-06, 6.900889e-06,
       1.685941e-04, 7.406938e-05, 7.406938e-05, 6.015995e-05,
       1.369339e-04, 5.604977e-06, 5.604977e-06, 1.369339e-04,...
 DESIRED: array([7.406938e-05, 1.462522e-04, 1.685941e-04, 1.417735e-04,
       8.785608e-05, 3.627328e-05, 6.900889e-06, 8.981213e-04,
       1.773368e-03, 2.044271e-03, 1.719061e-03, 1.065291e-03,...
====== 2 failed, 529 passed, 1 xfailed, 14 warnings in 106.61s (0:01:46) =======

[kp992]

Thanks @mmcky for checking that. Are they failing on your local machine too?

[kp992]

Ah, I get it. My local machine was using numpy 1.26 but with the numpy 2.2.1, I start getting these errors.

[mmcky]

thanks @kp992 - it will be interesting to start benchmarking the wasm and non-wasm versions via the test cases :-)

[mmcky]

@kp992 does this need to be numpy >= 2?

[kp992]

yeah, thanks. Nice catch!

[mmcky]

@kp992 happy for me to merge this? or want to make that change first?

[kp992]

sure, lets merge this.