The quality factor is now defined by
$$\bbox[10px,border:1px solid grey]{Q_p = {R \over X_{L_{p}}} \approx { R_s || Q^2_l R_l \over X_L}}$$
Quite obviously, therefore, $R_s$ does have an impact on the quality
factor of the network and the shape of the resonant curves.
If an ideal current source ($R_s = \infty Ω$) is employed, or if $R_s >> R_p$,
$$Q_p \approx { R_s || Q^2_l R_l \over X_L} = {Q^2_l R_l \over X_L} = {Q^2_l \over X_L / R_l} = {Q^2_l \over Q_l}$$
and
$$\bbox[10px,border:1px solid grey]{Q_p \approx Q_l} \,\, (Q_l \geq 10, Rs >> R_p)$$
Do you want to say or ask something?