Add compiler_verbose config flag

[ricardoV94]

It makes it easier for users to let us know what kind of graph isn't actually compiling this way.

---

📚 Documentation preview 📚: https://pytensor--1520.org.readthedocs.build/en/1520/

[codecov]

Codecov Report

Attention: Patch coverage is 50.00000% with 2 lines in your changes missing coverage. Please review.

Project coverage is 82.04%. Comparing base (ae38884) to head (82ab6cf).

Files with missing lines	Patch %	Lines
pytensor/link/numba/dispatch/basic.py	0.00%	1 Missing and 1 partial ⚠️

❌ Your patch check has failed because the patch coverage (50.00%) is below the target coverage (100.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1520      +/-   ##
==========================================
- Coverage   82.04%   82.04%   -0.01%     
==========================================
  Files         231      231              
  Lines       52364    52368       +4     
  Branches     9217     9218       +1     
==========================================
+ Hits        42962    42964       +2     
- Misses       7094     7095       +1     
- Partials     2308     2309       +1     

Files with missing lines	Coverage Δ
pytensor/configdefaults.py	73.63% <100.00%> (+0.08%)	⬆️
pytensor/configparser.py	92.58% <100.00%> (+0.02%)	⬆️
pytensor/link/numba/dispatch/basic.py	78.66% <0.00%> (-0.43%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[jessegrabowski]

Looks great, but need to add a section to docs/library/config.rst

[ricardoV94]

No I shouldn't, that should be generated automatically. It's bad design if it's not and I would open an issue for that

[ricardoV94]

It's the year 2025 ffs

[jessegrabowski]

In that case could you beef up the help string of the flag a bit, so that in the future when this section is auto-generated, it's helpful. It appears to only do anything in numba mode, and it dprints nodes? I somewhat don't get it.

[ricardoV94]

Right now it only does that but could be used for whenever we want to output information during compilation (as opposed to during rewrite). The node dprint is because when someone says numba is falling back to objectmode for stuff like AdvancedSubtensor that's not enough info, because there are 99 flavors of it

[jessegrabowski]

Fine then, I'm approving, but I will point out that you could have written something shorter than your reply to satisfy my request, rather than the reply itself.

[ricardoV94]

I'm AFK