[sim_costmodel](feat) auto scope is working on solve_tril

[kaixin1976]

1. auto scope is working on solve_tril and offer 12% perf improvement
2. sove_tril , fbgemm, gather_dot_min all corrected evaluated .
3. migrate auto blockify v1 from NPUIR to TA, so that cost model could evaluate cost based on blockified loops ,
   scope local super block is able to work based on the loops .
4. add in row-coalescing mechanism as part of layout optimization , so that FBGEMM could reproduce historical best performance record .
5. add 3 UT case to CI to guard the perforce of cost model and simd_simt mixed mode compiling .
6. remove duplicated / Invalid code to make the code easier to read , now effective code lines of simt cost model is below 4997, physically is still 5885 lines , still need to be reduced .
7. support delayed GSS
8. support sublocks
9. removed phase concept

New contributor declaration

• I am not making a trivial change, such as fixing a typo in a comment.

• I have written a PR description following these
  rules.

• I have run pre-commit run --from-ref origin/main --to-ref HEAD.

• Select one of the following.

  • I have added tests.
    • /test for lit tests
    • /unittest for C++ tests
    • /python/test for end-to-end tests
  • This PR does not need a test because FILL THIS IN.

• Select one of the following.

  • I have not added any lit tests.
  • The lit tests I have added follow these best practices,
    including the "tests should be minimal" section. (Usually running Python code
    and using the instructions it generates is not minimal.)

[kaixin1976]

/retry