Switch to dynamic_cast instead of string comparison for product retrieval

[knoepfel]

Dynamic casts were originally avoided when retrieving data products due to various RTTI difficulties across plugins. This PR restores them.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

❌ Your project status has failed because the head coverage (76.58%) is below the target coverage (80.00%). You can increase the head coverage or adjust the target coverage.

@@            Coverage Diff             @@
##             main     #203      +/-   ##
==========================================
- Coverage   76.61%   76.58%   -0.04%     
==========================================
  Files         124      124              
  Lines        2733     2733              
  Branches      475      475              
==========================================
- Hits         2094     2093       -1     
  Misses        446      446              
- Partials      193      194       +1     

Flag	Coverage Δ
unittests	76.58% <100.00%> (-0.04%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
phlex/model/products.hpp	95.00% <100.00%> (ø)

... and 5 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 14c7635...dc46c5c. Read the comment docs.

🚀 New features to boost your workflow:

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

[pcanal]

LGTM. (Although it is unclear what changed between the time it did not work and now).

[knoepfel]

LGTM. (Although it is unclear what changed between the time it did not work and now).

The last time I ran into the dynamic_cast problem was with older versions of apple-clang and GCC. I'm hopeful we're using new enough versions that the RTTI is consistent across dynamically loaded libraries. It seems to be consistent enough in the CI, at least (and on my Mac).