chore(reforcexy): add dynamic potential based rewards (#6)

chore(reforcexy): add dynamic potential based rewards (#6)

* chore(reforcexy): add dynamic potential based rewards

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): align tunables namespace

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): factor out pnl_target validation

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): cleanup PBRS integration

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): factor out default idle duration multiplier

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* chore: cleanup variable initialization

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor(reforcexy): factor out PBRS next state definition

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* chore: revert copilot incorrect suggestions

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): PBRS integration code cleanups

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): handle overtflow

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor(reforcexy): avoid code duplication

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor(reforcexy): add PBRS instrumentation

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): factor out PBRS signal computation

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): add check to PBRS logic

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* chore: sync PBRS in RSA

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* refactor(reforcexy): cleanup PBRS integration in RSA

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* Apply suggestion from @Copilot

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* docs: add PBRS

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
* fix(reforcexy): ensure model gamma is transmitted to env

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
---------

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>