ref(native): Publish CRASHING early

[jpnurmi]

Publish an early CRASHING state from the crash handler right after saving the crash context to shared memory but before invoking the on_crash or before_send callbacks that execute arbitrary user code that may, in the worst case, crash the crash handler.

This not only hardens the flow for the upcoming crash-stack unwinding (#2174), but in general, allows the daemon to detect such situations and act accordingly, instead of the app vanishing without any notification, resulting in misleading "Parent process exited without crash" logs.

flowchart TB
    subgraph app["App / backend"]
        direction LR
        READY
        subgraph handler["Crash handler"]
            direction LR
            CRASHING["CRASHING (new)"] -->|"on_crash/before_send"| CRASHED
        end
        READY --> handler
    end
    subgraph daemon["Crash daemon"]
        direction LR
        PROCESSING --> PROCESSED --> CAPTURED --> DONE
    end
    app --> daemon
    
    style CRASHING stroke:red,stroke-width:2px

Loading

[codecov]

Codecov Report

❌ Patch coverage is 80.85106% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.39%. Comparing base (1ae39d3) to head (98293f1).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2180      +/-   ##
==========================================
- Coverage   75.48%   75.39%   -0.10%     
==========================================
  Files         103      103              
  Lines       28089    28119      +30     
  Branches     5150     5158       +8     
==========================================
- Hits        21202    21199       -3     
- Misses       5545     5587      +42     
+ Partials     1342     1333       -9     

🚀 New features to boost your workflow:

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